Lithium-ion transfer at the interfaces between LiCoO2 and LiMn2O4 thin film electrodes and organic electrolytes

Lithium-ion transfer at the interfaces between LiCoO2 and LiMn2O4 thin film electrodes and organic electrolytes

Journal of Power Sources 294 (2015) 460e464 Contents lists available at ScienceDirect Journal of Power Sources journal homepage: www.elsevier.com/lo...

721KB Sizes 0 Downloads 34 Views

Journal of Power Sources 294 (2015) 460e464

Contents lists available at ScienceDirect

Journal of Power Sources journal homepage: www.elsevier.com/locate/jpowsour

Lithium-ion transfer at the interfaces between LiCoO2 and LiMn2O4 thin film electrodes and organic electrolytes Izumi Yamada, Kohei Miyazaki, Tomokazu Fukutsuka, Yasutoshi Iriyama, Takeshi Abe*, Zempachi Ogumi Graduate School of Engineering, Kyoto University, Nishikyo-ku, Kyoto 615-8510, Japan

h i g h l i g h t s  Interfacial Liþ transfer resistances were studied using thin film electrodes.  Large activation barriers (about 50 kJ mol1) on LiMn2O4 and LiCoO2 were observed.  The resistances of LiMn2O4 and LiCoO2 showed different changes with cycles.

a r t i c l e i n f o

a b s t r a c t

Article history: Received 13 March 2015 Received in revised form 9 June 2015 Accepted 17 June 2015 Available online xxx

Interfacial reactions at positive electrodes and organic electrolytes interface for Li-ion batteries are studied by AC impedance methods. Thin film electrodes with flat and smooth surface are prepared by pulsed laser deposition, and the Li-ion transfer is investigated at structurally ordered interface. Charge transfer resistances attributed to Li-ion transfer at the interface are observed. The charge transfer resistances on LiMn2O4 thin film electrode are much smaller than those on c-axis orientated LiCoO2 thin film electrode, indicating that the charge transfer resistances are influenced by the number of active sites at the interface. Irrespective with positive electrode materials, activation energies evaluated from the temperature dependence of Li-ion transfer resistances are almost similar (about 50 kJ mol1). These large activation energies suggest the existence of large energy barrier for Li-ion transfer at the interface between positive electrodes and organic electrolytes. After several chargeedischarge cycles, some differences in the charge transfer resistances are observed; the increases in resistances on LiMn2O4 thin film electrode are smaller than those on LiCoO2 thin film electrode. © 2015 Elsevier B.V. All rights reserved.

Keywords: Lithium-ion batteries Positive electrodes Interfaces Charge transfer Electrolyte decomposition AC impedance

1. Introduction Due to the concerns of the lack of energy sources and environmental pollution, secondary batteries, in particular Li-ion batteries have attracted much attention. Li-ion batteries have been used for portable electronic devices such as cellular phones and laptop personal computers, since Li-ion batteries have high cell potential and large gravimetric density. In addition, Li-ion batteries are recently utilized as a power supply for electronic vehicles (EV) and hybrid electronic vehicles (HEV). One of the most important properties required for Li-ion batteries in HEV is high rate performance, namely, capability for fast

* Corresponding author. E-mail address: [email protected] (T. Abe). http://dx.doi.org/10.1016/j.jpowsour.2015.06.108 0378-7753/© 2015 Elsevier B.V. All rights reserved.

charge and discharge reactions. Rate performances of Li-ion batteries are mainly governed by the reaction rate involved with Li-ion transfer. Therefore, kinetics of Li-ion transfer in batteries should be clearly understood. Li-ion transfer in Li-ion batteries includes three different steps: Li-ion transfer across the electrolyte/electrode interfaces, Li-ion diffusion through electrode, and Li-ion transport through electrolytes. Latter two steps have been well investigated so far in the research of solid-state physics and solution chemistry [1e5]. In addition to decrease the internal resistance of Li-ion batteries, diffusion path through active electrode material can be shortened by using smaller size particles [6], and electrolyte resistances can be decreased by using thinner electrolyte layers [7]. Then, it remains the case that the interfacial reaction should play an important role on the total rate of battery reactions. However, the interfacial Li-ion transfer has not been well understood yet. For elucidation of interfacial Li-ion transfer at an interface,

I. Yamada et al. / Journal of Power Sources 294 (2015) 460e464

461

structurally ordered interface of electrode is essential. Using flat and homogeneous thin film electrodes allows us to study the interfacial Li-ion transfer without considering the contributions of binder or conductive additives to electrochemical properties. In this paper, we have prepared LiCoO2 thin films and LiMn2O4 thin films by pulsed laser deposition (PLD), and then Li-ion transfer at electrode/electrolyte interface was studied. 2. Experimental Positive thin film electrodes of LiCoO2 and LiMn2O4 were prepared by PLD using KrF eximer laser of a wavelength of 248 nm (Japan Storage Battery, EXL-210) with polished Pt plate as a substrate. During deposition, a substrate was kept at 873 K for LiCoO2 and 973 K for LiMn2O4. Detailed deposition conditions were reported elsewhere [8,9]. Prepared thin films were characterized by X-ray diffraction (XRD) measurement using RINT-2500 (Rigaku) equipped with a graphite monochromator and a scintillation detector. Measurement conditions were 40 kV and 250 mA with a scanning speed of 0.125 s. Electrochemical properties of LiCoO2 and LiMn2O4 thin film electrodes were studied by cyclic voltammetry with the use of a three-electrode electrochemical cell by HSV-100 (Hokuto Denko Inc.). Lithium metal was used as reference and counter electrodes, and thin film electrodes as a working electrode (as shown in Fig. 1). In this study, all potentials are referred to Liþ/Li redox potential. An electrolyte used in this study was propylene carbonate (PC) containing 1 mol dm3 lithium perchlorate (LiClO4). Li-ion transfer at positive thin film electrodes was studied by AC impedance spectroscopy using Radiometer (Voltalab 40) over the frequency range from 100 kHz to 10 mHz with the same threeelectrode electrochemical cell. AC amplitudes were set at 10 mV. All experiments were conducted under Ar atmosphere. Contact surface areas of the electrodes were unified at 0.739 cm2 using Orings. 3. Results and discussion 3.1. Characterization of LiCoO2 Fig. 2a shows XRD pattern of a LiCoO2 thin film deposited on Pt plate at 873 K for 1 h. Number on peaks denotes index hkl. Except for substrate peaks derived from Pt, only one peak 2q ¼ 18.95 is

Fig. 2. X-ray diffraction patterns of (a) LiCoO2 and (b) LiMn2O4 thin films prepared on Pt plate. Number on peaks denotes index hkl.

remarkable indexed as the reflection of 003 plane of hexagonal LiCoO2. The XRD result shows that the LiCoO2 thin film had a preferred c-axis orientation to the substrate surface. Electrochemical behavior of LiCoO2 thin film was examined by cyclic voltammetry in 1 mol dm3 LiClO4/PC. As is shown in Fig. 3a, a large peak at 3.92 V and very small peaks at 4.08 and 4.18 V appeared in cathodic direction. These peaks correspond to phase transitions of LiCoO2 [10]. These results showed resultant LiCoO2 thin films showed almost similar electrochemical properties with LiCoO2 so far reported in literature [10]. The peak separation of the redox couple in Fig. 3a was about 40 mV. This small peak separation implies that the ohmic drop was small and, therefore, the electric conductivity of the LiCoO2 thin film was high enough for electrochemical measurements.

3.2. Characterization of LiMn2O4

Fig. 1. Schematic illustration of a three-electrode electrochemical cell.

Fig. 2b shows XRD pattern of LiMn2O4 thin film deposited on Pt substrate [11]. Several peaks were observed at 2q ¼ 18.5, 36.0, 38.0, 43.7, 57.8, 63.9, indexed as (111), (311), (222), (400), (511), (440), respectively (ICDD #35-0782). This XRD pattern indicates that single-phase spinel LiMn2O4 was obtained. Very sharp peaks

462

I. Yamada et al. / Journal of Power Sources 294 (2015) 460e464

Hence, it is reasonable to consider that the resistance was ascribed to the Li-ion (charge) transfer at the interface. Similarly to LiCoO2 thin film electrode/electrolyte interface, AC impedance measurements were conducted on LiMn2O4 thin film electrode. In Fig. 4b, Nyquist plots are shown at given potentials ranging 3.6e4.2 V. At an electrode potential of 3.6 V, no semicircle appeared and blocking-electrode-type behavior was observed, indicating no lithium ion insertion and extraction occurred on LiMn2O4 thin film electrode at this potential. This phenomenon is proved by the cyclic voltammogram given in Fig. 3b, in which no oxidation and reduction currents were observed at 3.6 V. At potentials above 3.84 V, one semicircle in the frequency region and linear line in the lower frequency appeared. Resistances evaluated from semicircles were dependent on electrode potentials, indicating that the impedance should be ascribed to the resistances of interfacial Li-ion transfer on LiMn2O4 thin films. Area-specific Nyquist plots are also shown in Fig. 4c and d. Characteristic frequencies were ca. 12.5 Hz for all Nyquist plots. Activation barrier at electrode/electrolyte interface for Li-ion transfer was studied. Fig. 5 shows temperature dependency of Liion transfer resistances of LiCoO2 (open circles) and LiMn2O4 (closed squares) in 1 mol dm3 LiClO4/PC. By the least-squares method, apparent activation energies for Li-ion transfer resistances were determined to be 46 ± 7.3 kJ mol1 for LiCoO2 at 4.10 V and 50 ± 2.4 kJ mol1 for LiMn2O4 at 3.94 V. The values of apparent activation energies were similar irrespective with positive electrode materials. These values are very large as compared with those for Liþ ion diffusion through active materials and Li-ion transport through liquid electrolytes [1,13]. Thus, these results indicate that a high activation barrier exists at the interface between positive electrode/electrolyte. Our recent studies for the large activation barriers at interface between electrode/electrolyte revealed that the de-solvation process was responsible for the activation energies [14e16]. 3.4. Li-ion transfer resistances Fig. 3. Cyclic voltammograms of (a) LiCoO2 and (b) LiMn2O4 thin film electrodes in 1 mol dm3 LiClO4/PC. Scanning rate was 0.1 mV s1.

demonstrate the resultant LiMn2O4 thin film was well crystallized. Redox behavior of LiMn2O4 thin film electrode was examined by cyclic voltammetry in 1 mol dm3 LiClO4/PC. As is shown in Fig. 3b, two couples of redox peaks were observed at 4.0 and 4.1 V. The peak separation was about 30 mV, indicating that the ohmic drop was very small as mentioned above, and therefore the electric conductivity of the LiMn2O4 thin film electrode was considered to be high enough for electrochemical measurements. 3.3. AC impedance spectroscopy AC impedance measurements were carried out at thin film electrode/electrolyte interface. In Fig. 4a, Nyquist plots of LiCoO2 thin film electrodes are shown at various potentials. At potential below 3.8 V, no semicircle but only blocking electrode behavior appeared. Little oxidation and reduction currents were observed in cyclic voltammogram in Fig. 3a below 3.8 V, and therefore blocking electrode behaviors were quite reasonable because few Liþ ions were inserted or extracted. At potentials above 3.88 V, one semicircle appeared in the higher frequency region followed by a diffusional behavior in the lower frequency region. Resistances evaluated from the semicircle were dependent on electrode potentials and also on the salt concentration of electrolyte used [12], indicating that the process corresponding to the observed semicircle was closely related with the relaxation process of Li ions.

As described before, one semicircle in the Nyquist plots should be ascribed to Li-ion transfer resistances. Fig. 6 shows the Li-ion transfer resistances as a function of electrode potentials in 1 mol dm3 LiClO4/PC on LiCoO2 thin film (open circles) and LiMn2O4 thin film (closed squares). Li-ion transfer resistances of LiCoO2 and LiMn2O4 exhibited similar dependency for the electrode potentials. The resistances were kept to minimum values at around 4.0 and 3.95 V for LiCoO2 and LiMn2O4, respectively. It should be noted that minimum values of Li-ion transfer resistances were largely different among positive electrode materials. At 4.0 V, the value of resistance on LiMn2O4 thin film electrode was 130 U; while that of LiCoO2 was 500 U, which is four times larger than that of LiMn2O4. Since the reactive surface areas were regulated to be the same size, this large difference should be caused by electrodes themselves. LiCoO2 has a layered 2D structure, and LiMn2O4 has a spinel-type 3D structure. In this study, we obtained the horizontally aligned LiCoO2 layers, and therefore, the active sites for Li-ion insertion and extraction were limited only to the defects of layers. In contrast, LiMn2O4 would have many sites for Liion insertion and extraction on its surface. The large difference in the charge transfer resistances should be related to the number of sites at the electrode/electrolyte interface. Fig. 7a shows the changes in the Li-ion transfer resistances of LiCoO2 as a function of electrode potentials during three chargeedischarge cycles. Each potential was maintained for 1 h in order to reach thermodynamic equilibrium states, prior to AC impedance measurements. In the charging direction, Li-ion transfer resistances decreased with increasing electrode potentials and then kept

Fig. 4. Impedance spectra of (a, c) LiCoO2 and (b, d) LiMn2O4 thin film electrodes in 1 mol dm3 LiClO4/PC. Red symbols denotes impedance plots at the frequency of 10n Hz (n: integer). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

constant with increasing electrode potential. In contrast, in the discharging direction, Li-ion transfer resistances increased with decreasing electrode potentials. Although this tendency remained unchanged, the absolute value of resistances increased with an increase in cycles. Therefore, it can be said that the Li-ion transfer resistances were enlarged when the electrode potentials were higher than ca. 4.1 V, and the resistances showed almost the same curves (1st discharge and 2nd charge, and 2nd discharge and 3rd charge) at the potentials below 4.1 V. This potential transition implies that some irreversible reactions occurred and a passivation film should be formed on LiCoO2 thin film electrode at higher electrode potential [17].

Fig. 5. (a) Temperature dependence of charge transfer resistances of positive thin film/ electrolyte interface. Lines are the result of a least-squares method. (b) An equivalent circuit for analyzing Nyquist plots.

Fig. 6. Charge transfer resistances as a function of the electrode potentials.

464

I. Yamada et al. / Journal of Power Sources 294 (2015) 460e464

observed for LiCoO2 and LiMn2O4 thin film electrodes in terms of reversibility at higher electrode potentials. The irreversible behavior of LiCoO2 can be ascribed to the electrolyte decomposition. Electrolyte decomposition might be catalytically promoted by transition metals of positive electrode materials, and therefore the different behavior of Li-transfer resistances appeared between LiCoO2 and LiMn2O4 electrodes. 4. Conclusions Li-ion transfer at the interface between positive electrodes and organic electrolytes was investigated, using structurally ordered surfaces of LiCoO2 and LiMn2O4 thin film electrodes prepared by PLD. It is found that Li-ion transfer resistances at the interfaces were influenced by the structures of positive electrode materials: caxis oriented LiCoO2 thin films had larger Li-ion transfer resistances, and spinel-type LiMn2O4 smaller resistances. Temperature dependency of Li-ion transfer resistances displayed that there were large activation barriers at the interface between positive electrodes and organic electrolytes. As the chargeedischarge cycles proceeded, irreversible behavior of Li-ion transfer resistances on LiCoO2 thin films appeared, but LiMn2O4 showed little change in resistances during chargeedischarge cycles. These differences could be ascribed to the different reactivity of electrode materials for the electrolyte decomposition. References

Fig. 7. Charge transfer resistances at (a) LiCoO2 and (b) LiMn2O4 thin film electrodes as a function of the electrode potentials.

In the case of LiMn2O4 (Fig. 7b), the changes in Li-ion transfer resistances were smaller than those on LiCoO2, both in the charge and discharge directions expect for the first charge. Differently from the case of LiCoO2, Li-ion transfer resistances decreased with decreasing electrode potentials, and the absolute value of resistances did not remarkably increase. Different behaviors were

[1] A. Van der Ven, G. Ceder, Electrochem. Solid State Lett. 3 (2000) 301. [2] D. Aurbach, K. Gamolsky, B. Markovsky, G. Salitra, Y. Gofer, U. Heider, R. Oesten, M. Schmidt, J. Electrochem. Soc. 147 (2000) 1322. [3] A. Funabiki, M. Inaba, Z. Ogumi, S. Yuasa, J. Otsuji, A. Tasaka, J. Electrochem. Soc. 145 (1998) 172. [4] A. Ferrya, G. Oradd, P. Jacobsson, J. Chem. Phys. 108 (1998) 7426. [5] B. Klassen, R. Aroca, M. Nazri, G.A. Nazri, J. Phys. Chem. B 102 (1998) 4795. [6] J.B. Bates, N.J. Dudney, B. Neudecker, A. Ueda, C.D. Evans, Solid State Ionics 135 (2000) 33. [7] B.J. Neudecker, N.J. Dudney, J.B. Bates, J. Electrochem. Soc. 147 (2000) 517. [8] Y. Iriyama, M. Inaba, T. Abe, Z. Ogumi, J. Power Sources 94 (2001) 175. [9] M. Inaba, T. Doi, Y. Iriyama, T. Abe, Z. Ogumi, J. Power Sources 81e82 (1999) 554. [10] J.N. Reimers, J.R. Dahn, J. Electrochem. Soc. 139 (1992) 2091. [11] I. Yamada, T. Abe, Y. Iriyama, Z. Ogumi, Electrochem. Comm. 5 (2003) 502. [12] I. Yamada, Y. Iriyama, T. Abe, Z. Ogumi, J. Power Sources 172 (2007) 933. [13] G. Kumar, Janaklraman, N. Namboodlrl, Gangadharan, Kanagaraj, Saminathan, N. Sharlef, J. Chem. Eng. Data 36 (1991) 467. [14] T. Abe, H. Fukuda, Y. Iriyama, Z. Ogumi, J. Electrochem. Soc. 151 (2004) A1120. [15] F. Sagane, T. Abe, Y. Iriyama, Z. Ogumi, J. Power Sources 146 (2005) 749. [16] T. Doi, K. Miyatake, Y. Iriyama, T. Abe, Z. Ogumi, T. Nishizawa, Carbon 42 (2004) 3183. [17] D. Aurbach, J. Power Sources 89 (2000) 206.