- Email: [email protected]
Structural changes and electrochemical properties of Li2Cu1 − xMxO2 for lithium secondary batteries
SOSI-13189; No of Pages 5 Solid State Ionics xxx (2013) xxx–xxx
Contents lists available at ScienceDirect
Solid State Ionics journal homepage: www.elsevier.com/locate/ssi
Structural changes and electrochemical properties of Li2Cu1 − xMxO2 for lithium secondary batteries Elly Setiawati, Masahiko Hayashi ⁎, Masayuki Tsuda, Katsuya Hayashi, Ryuichi Kobayashi NTT Energy and Environment Systems Laboratories, Nippon Telegraph and Telephone (NTT) Corporation, Morinosato Wakamiya 3-1, Atsugi, Kanagawa 243-0198, Japan
a r t i c l e
i n f o
Article history: Received 20 May 2013 Received in revised form 8 November 2013 Accepted 13 November 2013 Available online xxxx Keywords: Lithium copper oxide Li2CuO2 Cathode Cation substitution Irreversible capacity Structural change
a b s t r a c t Li2CuO2 substituted by various cations, such as Co, Ni, Fe, Mn, and Ti, was examined to decrease the irreversible capacity of unsubstituted one. Among several kinds of cations tested, Ni and Co were found to be effective in improving the irreversible capacity. Li2Cu0.7Ni0.3O2 exhibited a reversible capacity of about 150 mAh/g, while the unsubstituted one exhibited a capacity of only about 100 mAh/g. The structural changes in the Cu-based oxides that occurred during the electrochemical reaction were investigated with ex-situ XRD measurements. The results indicated that the structural transformation behavior changed as a result of the cation substitution. © 2013 Published by Elsevier B.V.
1. Introduction Lithium secondary batteries including rocking-chair type lithium ion batteries (LIBs) are widely used as power sources for electronic tools such as cell phones and laptop computers. The use of large scale LIBs has spread to power sources for electric vehicles and energy storage tools for smart grid systems combined with recyclable energy such as solar and wind power [1,2]. The price of LIBs should be low, especially for large-scale utilization. However, most of the cathode materials used for the LIBs contain rare metals such as Mn, Ni, Ti and Co [3,4]. It is essential to greatly decrease the use of rare metals other than lithium. Cu-based oxides such as LiCuO2, Li2CuO2 and NaCuO2 have already been reported as rare metal-free cathode materials [5–12]. Copper element is a common and inexpensive metal element without concern on natural resource. These compounds have one-dimensional [CuO4] square chains, where copper is at the center of an oxygen rectangle. The [CuO4] chains in Li2CuO2 are located in a corrugated manner. This can be regarded as a layer-like structure, which enables smooth Li-ion diffusion in the oxide. Moreover, Li2CuO2 has a high theoretical capacity of 245 mAh/g corresponding to the following electrochemical reaction: Li2CuO2 → LiCuO2 + Li+ + e−. The electrochemical behavior of Li2CuO2 has been reported in detail by Arai et al. [8]. The Li2CuO2 (space group: Immm) was electrochemically oxidized to Li1.5CuO2 ⁎ Corresponding author. Tel.: +81 46 240 3760; fax: +81 46 270 3721. E-mail address: [email protected] (M. Hayashi).
(C2/m) and it was finally oxidized to LiCuO2 (C2/m) during charging. The first charge and discharge capacities were 250 and 130 mAh/g, respectively. The reversible capacity was 130 mAh/g after the second cycle. This reversible capacity corresponds to the reversible reaction between Li1.5CuO2 and LiCuO2. The large irreversible capacity at the first cycle occurred so that the final discharge product, LiCuO2, could not be electrochemically reduced to Li1.5CuO2 during discharging. This irreversible reaction might be due to relatively large structural changes in Li2CuO2, Li1.5CuO2 and LiCuO2, although all these oxides have onedimensional [CuO4] chains. Furthermore, Imanishi et al. reported that the first irreversible capacity was suppressed by substituting the Cu sites of Li2CuO2 by Ni and that Li2Cu0.6Ni0.4O2 exhibited a maximum first discharge capacity of about 320 mAh/g [9]. Love et al. also reported that Li2Cu0.5Ni0.4M0.1O2 (M = Al, Ga) prepared by micro emulsion method exhibits better cyclability and capacity retention than unsubstituted Li2CuO2 [12]. These results indicate that the partial substitution of Cu sites is an effective way of improving the reversibility of Li2CuO2. There are no reports on the electrochemical properties of Li2CuO2 substituted by transition metals other than Ni. It is important to investigate effects of substitution with various kinds of metals to reduce the irreversible capacity of Li2CuO2. In this study, we investigated the electrochemical properties and structural changes in Li2CuO2 substituted by various transition metal ions such as Co, Ni, Fe, Ti, and Mn, and examined its effect in suppressing the irreversible capacity at the first cycle and increasing the reversible capacity.
0167-2738/$ – see front matter © 2013 Published by Elsevier B.V. http://dx.doi.org/10.1016/j.ssi.2013.11.043
Please cite this article as: E. Setiawati, et al., Solid State Ionics (2013), http://dx.doi.org/10.1016/j.ssi.2013.11.043
2
E. Setiawati et al. / Solid State Ionics xxx (2013) xxx–xxx
2. Experimental
CuKα
2.1. Synthesis of Li2Cu1 − xMxO2 (M = Co, Ni, Fe, Mn, and Ti)
(f) Intensity (a.u.)
Lithium copper oxide (Li2CuO2) was synthesized by a simple solid state reaction involving a lithium source [lithium hydroxide monohydrate (LiOH·H2O, Kanto Chemical Co., Inc.)] and a high purity copper source [copper oxide (CuO, Kanto Chemical Co., Inc.)] at 850 °C for 24 h under O2 gas flow. The resultant powder was then stored in a dry room with a dew point of less than −40 °C. The same procedure was used for the cation, M (M = Co, Ni, Fe, Ti, and Mn), -substituted oxides. Mono oxides (Kanto Chemical Co., Inc.) for M = Co, Ni, Mn and metal powders (Kanto Chemical Co., Inc.) for M = Fe and Ti were used as cation sources. The crystallographic structure of the resultant powder was characterized with an X-ray diffractometer (Rigaku, RINT2500) using CuKα radiation under a constant power of 30 kV and 100 mA. The formation phases of the synthesized powder were identified using the database of the International Center for Diffraction Data (ICCD). The particle morphology was observed using a scanning electron microscope (SEM, SU1510, Hitachi High-Technologies Corp.).
(e) (d) (c) (b) (a) Li CuO(PDF:01-079-1975) 2
2.2. Electrochemical measurements of Li2Cu1 − xMxO2 (M = Co, Ni, Fe, Mn, and Ti) Electrochemical measurements were carried out using a twoelectrode coin-type test cell (type 2320). An electrode material mixture was prepared with Cu-based oxide (70 wt.%), Ketjen Black EC600JD (25 wt.%, Lion Co., Ltd.), and polytetrafluoroethylene (5 wt.%, Daikin Co., Ltd.). The mixture was then roll-pressed into a 0.5 mm thick sheet. This sheet was cut into a circle shape and used as a working electrode in a coin-type test cell. Lithium metal (0.3 mm thick) was used as a counter electrode. Microporous polypropylene film (Celgard), which was located between the two electrodes, was used as a separator. The electrolyte was 1 mol/L of LiPF6 in an equal volume of ethylene carbonate (EC) and dimethyl carbonate (DMC) solution (Kishida Chemical Co., Ltd.). The entire preparation and assembly process was carried out in a dry room with dew point of less than −40 °C. The tests on the coin-type cell were conducted in a constant temperature-controlled chamber maintained at 20 °C. The electrochemical measurement was carried out at a constant current density of 1 mA/cm2 in a voltage range of 2.0–4.0 V, with a rest period of 10 min between each charge and discharge step. 2.3. Stability evaluation of Li2Cu1 − xMxO2 (M = Co, Ni, Fe, Mn, and Ti) An ex-situ evaluation of the discharged/charged electrodes containing the Cu-based oxides was carried out to study the stability of the crystal structure when it underwent an electrochemical process. Test cells were performed under the same condition as the electrochemical evaluation (current density: 1 mA/cm2) but they were terminated under different conditions on the cut-off voltages or discharge/charge capacities. After the electrochemical evaluation, the cells were disassembled and the pellet electrode containing the Cu-based oxide was removed from the cell and washed several times with dimethylcarbonate solution (DMC, Kishida Chemical Co., Ltd.) to eliminate the lithium salt particles in the electrode. The pellet electrodes were dried and stored in a vacuum condition prior to the XRD measurements. 3. Results and discussion Fig. 1 shows XRD patterns of Li2Cu0.9M0.1O2 [M = Cu (unsubstituted), Co, Ni, Fe, Mn, and Ti]. Almost all the peaks for the cation-substituted Li2CuO2 corresponded to PDF data #01-079-1975 (space group: Immm) for Li2CuO2, although there were some new peaks or peak shifts in the patterns. No peaks indexed to the raw materials of monoxides or metal
15
20
25
2
30
35
40
2θ (deg) Fig. 1. XRD patterns of Li2Cu0.9M0.1O2 [M = (a) Cu (unsubstituted), (b) Co, (c) Ni, (d) Fe, (e) Ti, and (f) Mn].
were confirmed in the patterns. These results indicate that the cations can substitute the Cu sites in Li2CuO2. Morphology changes in the cation-substituted Li2CuO2 were analyzed by SEM observation. Fig. 2 shows SEM images of Li2Cu0.9Ni0.1O2 as a typical cation-substituted Cu-based oxide compared with Li2CuO2. The Nisubstituted Li2CuO2 particles are more homogeneous and have a smooth shape without a sharp edge compared with Li2CuO2. This result suggests that the crystallization temperature became lower as a result of the cation substitution, and the sintering of the particles can proceed easily. Fig. 3 shows the first charge and discharge curves of Li2Cu0.9M0.1O2 (M = Co, Ni, Fe, Ti, and Mn) electrodes compared with those of the unsubstituted Li2CuO2 electrodes. All the electrodes exhibited charge and discharge capacities of about 240 mAh/g and about 130 mAh/g, respectively. These values correspond to about 1.0 mol and about 0.6 mol equivalent of lithium in Li2Cu1 − xMxO2. Moreover, irreversible capacities of about 110 mAh/g were observed in all oxides, regardless of the substituent metals. As for Li2CuO2, this first discharge capacity agreed well with that of 130 mAh/g reported by Arai et al. [8]. Consequently, no clear improvement in the irreversible capacity could be achieved with a cation-substitution content, x = 0.1. However, it can also be seen that Ni- and Co-substituted Li2CuO2 electrodes have slightly larger discharge capacities than unsubstituted Li2CuO2. Fig. 4 shows the cycle properties of Li2Cu0.9M0.1O2 (M = Co, Ni, Fe, Ti, and Mn) electrodes compared with those of the unsubstituted Li2CuO2 electrodes. As shown in this figure, compared with the unsubstituted oxide, the Fe- and Mn-substituted oxides exhibited smaller discharge capacities and that of the Ti-substituted oxide was similar. On the other hand, the Ni- and Co-substituted oxides exhibited the larger discharge capacities than the unsubstituted oxide. In addition, the discharge capacities of the Ni-substituted oxide gradually increased as the cycle number increased. This unique behavior was not observed for any other cation-substituted oxide other than Ni. However, similar behavior has been observed in the cycle properties of Li2Cu0.8Ni0.2O2 as reported by Imanishi et al. [9], but they did not describe the reason for this behavior in detail. Fig. 5 shows the correlation between the Ni- and Co-substitution content, x, and the first charge and discharge capacities of Li2Cu1 − xMxO2
Please cite this article as: E. Setiawati, et al., Solid State Ionics (2013), http://dx.doi.org/10.1016/j.ssi.2013.11.043
E. Setiawati et al. / Solid State Ionics xxx (2013) xxx–xxx
(i)
3
(ii)
Fig. 2. SEM images of (i) Li2CuO2 and (ii) Li2Cu0.9Ni0.1O2.
(M = Co, Ni) electrodes. The charge capacities for both Ni- and Cosubstituted oxides deceased as the substitution content, x, increased. The Co-substituted oxides exhibited similar discharge capacities of about 140 mAh/g for the tested substitution condition and the x = 0.7 oxide exhibited almost no irreversible capacity. On the other hand, the Ni-substituted oxide exhibited the largest discharge capacity of about 160 mAh/g at x = 0.3. The x = 0.9 oxide exhibited the smallest irreversible capacity during the substation condition tested, although the oxide at this composition exhibited the smallest discharge capacity of about 100 mAh/g. As mentioned above, the Co and Ni substitution led to a significant decrease in the irreversible capacity. Such substitution of Cu sites by Co and Ni might suppress the transformation of the pristine Li2CuO2 structure, which was followed by a decrease in the irreversible capacity. As for Li2CuO2 and Ni-substituted oxides, the tendency on electrochemical properties agreed roughly with the previous reports [8–12]. However, the values of the charge/discharge capacities disagree somewhat with them. This may be due to the difference in the heat treatment condition. In this study, the heat treatment was carried out under O2 gas flow to enhance the cation-substitution reaction, whereas it was carried out in air [8,10] or in an Ar stream [9] in the previous reports. This difference in the calcination condition may have some effect on the oxidation state of Cu and Ni. Consequently, our results did not agree completely with those of the previous reports. Li2Cu0.1Ni0.9O2 and Li2Cu0.3Co0.7O2, which contained large amounts of rare metals such as Ni or Co, exhibited the smallest irreversible capacities. However, the properties of Li2Cu0.7Ni0.3O2 and Li2Cu0.9Co0.1O2, which contained less rare metals, were intensively investigated. These two oxides had characteristic properties. The former oxide exhibited the largest discharge capacity, and the latter exhibited the largest charge capacity. Fig. 6 shows the cycle properties of batteries with
4
Discharge capacity (mAh/g)
Voltage (V)
200
(f) (c)
3.5
(b)
(d) (a) (e) 3
(b) (f) 2.5
(e) (d)
2
Li2Cu0.7Ni0.3O2 and Li2Cu0.9Co0.1O2 cathode materials. Li2Cu0.9Co0.1O2 exhibited a stable capacity of about 130 mAh/g. On the other hand, the discharge capacity of Li2Cu0.7Ni0.3O2 gradually increased from 160 to 210 mAh/g as shown in Fig. 6(ii). A similar result was seen with Li2Cu0.9Ni0.1O2 as shown in Fig. 4(c). Of the oxides examined, the Nisubstituted Li2CuO2 showed this gradual increase in discharge capacity. The structural changes in the Cu-based oxides during the charge and discharge process were examined to understand the effect of substituting the Cu sites of Li2CuO2 by Ni and Co. Fig. 7 shows the XRD patterns of assynthesized Cu-based oxide powder and ex-situ XRD patterns of the electrodes containing the oxides after half charging (charge capacity: 125 mAh/g), full charging (cut-off voltage: 4.0 V), and full discharging (cut-off voltage: 2.0 V) at the first cycle. The unsubstituted Li2CuO2 structure transformed into LiCuO2 via an intermediate Li1.5CuO2 structure with a trace of Li2CuO2 during charging and then Li1.5CuO2 and Li2CuO2 coexist after discharging as shown in Fig. 7(i). The traces of Li2CuO2 that exist after full charging and full discharging are thought to be unreacted parts in the pristine material. After the subsequent cycle, the cycle seemed to be done between Li1.5CuO2 (discharging state) and Li1CuO2 (charging state) from a calculation of the discharge and charge capacities. These structural changes are consistent with those reported by Arai. The changes between the Li1.5CuO2 structure and the Li1CuO2 structure were also seen in the XRD patterns of the Co-substituted oxide. Therefore, these results suggest that the reaction mechanism for the Cosubstituted oxide would proceed in an almost similar manner to that for the unsubstituted oxide. As a result of the above-mentioned behavior, the unsubstituted and Co-substituted oxides exhibited almost the same irreversible capacities as shown in Fig. 3. However, there are differences between the unsubstituted and Co-substituted oxide as for the discharge products. After full discharging, the unsubstituted Li2CuO2 contained a mixed phase of Li2CuO2 and Li1.5CuO2, while the Li2Cu0.9Co0.1O2 contained only the Li1.5CuO2 phase. These results indicated that the
(c) (a)
150
100
0 0
50
100
150
200
250
300
Capacity (mAh/g) Fig. 3. First charge and discharge profiles for Li2Cu0.9M0.1O2 electrodes [M = (a) Cu (unsubstituted), (b) Co, (c) Ni, (d) Fe, (e) Ti, and (f) Mn].
: (a) Cu : (b) Co : (c) Ni
50
10
20
30
40
: (d) Fe : (e) Ti : (f) Mn 50
60
Cycle number Fig. 4. Cycle properties of Li2Cu0.9M0.1O2 electrodes [M = (a) Cu (unsubstituted), (b) Co, (c) Ni, (d) Fe, (e) Ti, and (f) Mn].
Please cite this article as: E. Setiawati, et al., Solid State Ionics (2013), http://dx.doi.org/10.1016/j.ssi.2013.11.043
4
E. Setiawati et al. / Solid State Ionics xxx (2013) xxx–xxx
CuKα
300
Capacity (mAh/g)
Charge
(i) Li CuO
: (i) Ni : (ii) Co
2
2
: Li2CuO2 (PDF #01-079-1975 ) : Li1.5CuO2 (PDF #01-079-1505 ) : Li1CuO2 (PDF #01-085-1926 )
250
200
(a) (b)
150
Discharge 100
0
0.2
0.4
(c) 0.6
0.8
1
(d)
x in Li2Cu1-xMxO2 Fig. 5. First charge and discharge capacities of Li2Cu1 − xMxO2 [M = (i) Ni, and (ii) Co] electrodes.
(ii) Li2Cu0.7Ni0.3O2
: Li2CuO2 (PDF #01-079-1975 ) : Li1.5CuO2 (PDF #01-079-1505 ) : Li1CuO2 (PDF #01-085-1926 ) : LiNiO2 (PDF #01-085-1966 )
(a) complete reversibility between Li2CuO2 and Li1.5CuO2 is not achieved by Co substitution. Moreover, Co substitution promotes the formation of layer-like structures Li1.5CuO2 and LiCuO2, and results in a slight increase in discharge capacity. On the other hand, structural changes between Li1.5CuO2 and Li1CuO2 structure were confirmed in the XRD patterns of Li2Cu0.7Ni0.3O2 shown in Fig. 7(ii). In addition, no peaks corresponding to the Li2CuO2 were found in the XRD patterns after full charging and full discharging as shown in Fig. 7(ii), (c) and (d). This result indicates that Ni-substitution promotes an electrochemical reaction between Li1.5CuO2 and LiCuO2 in a similar way with Co substitution. In addition to such changes, new peaks corresponding to LiNiO2 appeared at 2θ = 18.7 and 36.6°. To be precise, this phase should be noted as LixNiO2 because there were slight peak shifts in the XRD patterns seen in Fig. 7(ii), (b)–(d). These results indicate that the Ni-substituted oxide segregated two phases, LixCuO2 and LixNiO2. LixNiO2 is well-known as a cathode material for Li ion batteries and Li ions can be inserted/
(b) (c) (d)
(iii) Li2Cu0.9Co0.1O2
: Li2CuO2 (PDF #01-079-1975 ) : Li1.5CuO2 (PDF #01-079-1505 ) : Li1CuO2 (PDF #01-085-1926 )
(a) (b) 250
Discharge capacity (mAh/g)
(c) (d)
(i)
15
20
25
30
35
40
2θ θ (deg)
200
Fig. 7. XRD patterns of (a) as-synthesized Cu-based oxide powder and ex-situ XRD patterns of (b) half-charged, (c) charged, and (d) discharged electrodes containing Cubased oxides. (i) Li2CuO2 (unsubstituted), (ii) Li2Cu0.7Ni0.3O2 and (iii) Li2Cu0.9Co0.1O2.
150 extracted to/from this oxide. These results suggest that the formation of LixNiO2 is responsible for the characteristic behavior confirmed in the Ni-substituted oxides where there is a gradual increase in discharge capacities.
(ii)
100
4. Conclusions
5
10
15
20
25
Cycle number Fig. 6. Cycle properties of (i) Li2Cu0.7Ni0.3O2 and (ii) Li2Cu0.9Co0.1O2 electrodes.
Li2CuO2 exhibits two different electrochemical reactions; an irreversible one of Li2CuO2 to Li1.5CuO2 and a reversible one of Li1.5CuO2 to LiCuO2. Substituting Cu sites with transition metal ions was
Please cite this article as: E. Setiawati, et al., Solid State Ionics (2013), http://dx.doi.org/10.1016/j.ssi.2013.11.043
E. Setiawati et al. / Solid State Ionics xxx (2013) xxx–xxx
ineffective in improving the reversibility of Li2CuO2 and Li1.5CuO2, however Ni- and Co-substituted oxides showed a slight improvement in their reversible capacities. Both Ni and Co substitutions seemed to enhance the electrochemical reaction between Li1.5CuO2 and LiCuO2. Moreover, the formation of LixNiO2 in the Ni-substituted oxide caused a gradual increase in discharge capacity. Acknowledgments We thank Dr. Y. Sugiyama for his helpful guidance. We also thank Mr. H. Ota and Mr. T. Yamada for their help with the experiments.
5
References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12]
L. Lu, X. Han, J. Li, J. Hua, M. Ouyang, J. Power Sources 226 (2013) 272. C. Ahn, C. Li, H. Peng, J. Power Sources 196 (2011) 10369. M.S. Whittingham, Chem. Rev. 104 (2004) 4271. J.W. Fergus, J. Power Sources 195 (2010) 939. A.R. Wizansky, P.E. Rauch, F.J. Disalvo, J. Solid State Chem. 81 (1989) 203. K. Imai, M. Koike, H. Takei, H. Sawa, D. Shiomi, M. Kinoshita Nozawa, J. Phys. Soc. Jpn. 61 (1819) (1992). R. Berger, J. Less-Common Met. 169 (1991) 33. H. Arai, S. Okada, Y. Sakurai, J. Yamaki, Solid State Ionics 106 (1998) 45. N. Imanishi, K. Shizuka, T. Ikenishi, T. Matsumura, A. Hirano, Y. Takeda, Solid State Ionics 177 (2006) 1341. Y. Arachi, Y. Nakata, K. Hinoshita, T. Setsu, J. Power Sources 196 (2011) 6939. Y. Arachi, T. Setsu, T. Ide, K. Hinoshita, Y. Nakata, J. Power Sources 225 (2012) 611. C.T. Love, M.D. Johannes, A.M. Stux, K.E. Swider-Lyons, ECS Trans. 16 (29) (2009) 27.
Please cite this article as: E. Setiawati, et al., Solid State Ionics (2013), http://dx.doi.org/10.1016/j.ssi.2013.11.043
Contents lists available at ScienceDirect
Solid State Ionics journal homepage: www.elsevier.com/locate/ssi
Structural changes and electrochemical properties of Li2Cu1 − xMxO2 for lithium secondary batteries Elly Setiawati, Masahiko Hayashi ⁎, Masayuki Tsuda, Katsuya Hayashi, Ryuichi Kobayashi NTT Energy and Environment Systems Laboratories, Nippon Telegraph and Telephone (NTT) Corporation, Morinosato Wakamiya 3-1, Atsugi, Kanagawa 243-0198, Japan
a r t i c l e
i n f o
Article history: Received 20 May 2013 Received in revised form 8 November 2013 Accepted 13 November 2013 Available online xxxx Keywords: Lithium copper oxide Li2CuO2 Cathode Cation substitution Irreversible capacity Structural change
a b s t r a c t Li2CuO2 substituted by various cations, such as Co, Ni, Fe, Mn, and Ti, was examined to decrease the irreversible capacity of unsubstituted one. Among several kinds of cations tested, Ni and Co were found to be effective in improving the irreversible capacity. Li2Cu0.7Ni0.3O2 exhibited a reversible capacity of about 150 mAh/g, while the unsubstituted one exhibited a capacity of only about 100 mAh/g. The structural changes in the Cu-based oxides that occurred during the electrochemical reaction were investigated with ex-situ XRD measurements. The results indicated that the structural transformation behavior changed as a result of the cation substitution. © 2013 Published by Elsevier B.V.
1. Introduction Lithium secondary batteries including rocking-chair type lithium ion batteries (LIBs) are widely used as power sources for electronic tools such as cell phones and laptop computers. The use of large scale LIBs has spread to power sources for electric vehicles and energy storage tools for smart grid systems combined with recyclable energy such as solar and wind power [1,2]. The price of LIBs should be low, especially for large-scale utilization. However, most of the cathode materials used for the LIBs contain rare metals such as Mn, Ni, Ti and Co [3,4]. It is essential to greatly decrease the use of rare metals other than lithium. Cu-based oxides such as LiCuO2, Li2CuO2 and NaCuO2 have already been reported as rare metal-free cathode materials [5–12]. Copper element is a common and inexpensive metal element without concern on natural resource. These compounds have one-dimensional [CuO4] square chains, where copper is at the center of an oxygen rectangle. The [CuO4] chains in Li2CuO2 are located in a corrugated manner. This can be regarded as a layer-like structure, which enables smooth Li-ion diffusion in the oxide. Moreover, Li2CuO2 has a high theoretical capacity of 245 mAh/g corresponding to the following electrochemical reaction: Li2CuO2 → LiCuO2 + Li+ + e−. The electrochemical behavior of Li2CuO2 has been reported in detail by Arai et al. [8]. The Li2CuO2 (space group: Immm) was electrochemically oxidized to Li1.5CuO2 ⁎ Corresponding author. Tel.: +81 46 240 3760; fax: +81 46 270 3721. E-mail address: [email protected] (M. Hayashi).
(C2/m) and it was finally oxidized to LiCuO2 (C2/m) during charging. The first charge and discharge capacities were 250 and 130 mAh/g, respectively. The reversible capacity was 130 mAh/g after the second cycle. This reversible capacity corresponds to the reversible reaction between Li1.5CuO2 and LiCuO2. The large irreversible capacity at the first cycle occurred so that the final discharge product, LiCuO2, could not be electrochemically reduced to Li1.5CuO2 during discharging. This irreversible reaction might be due to relatively large structural changes in Li2CuO2, Li1.5CuO2 and LiCuO2, although all these oxides have onedimensional [CuO4] chains. Furthermore, Imanishi et al. reported that the first irreversible capacity was suppressed by substituting the Cu sites of Li2CuO2 by Ni and that Li2Cu0.6Ni0.4O2 exhibited a maximum first discharge capacity of about 320 mAh/g [9]. Love et al. also reported that Li2Cu0.5Ni0.4M0.1O2 (M = Al, Ga) prepared by micro emulsion method exhibits better cyclability and capacity retention than unsubstituted Li2CuO2 [12]. These results indicate that the partial substitution of Cu sites is an effective way of improving the reversibility of Li2CuO2. There are no reports on the electrochemical properties of Li2CuO2 substituted by transition metals other than Ni. It is important to investigate effects of substitution with various kinds of metals to reduce the irreversible capacity of Li2CuO2. In this study, we investigated the electrochemical properties and structural changes in Li2CuO2 substituted by various transition metal ions such as Co, Ni, Fe, Ti, and Mn, and examined its effect in suppressing the irreversible capacity at the first cycle and increasing the reversible capacity.
0167-2738/$ – see front matter © 2013 Published by Elsevier B.V. http://dx.doi.org/10.1016/j.ssi.2013.11.043
Please cite this article as: E. Setiawati, et al., Solid State Ionics (2013), http://dx.doi.org/10.1016/j.ssi.2013.11.043
2
E. Setiawati et al. / Solid State Ionics xxx (2013) xxx–xxx
2. Experimental
CuKα
2.1. Synthesis of Li2Cu1 − xMxO2 (M = Co, Ni, Fe, Mn, and Ti)
(f) Intensity (a.u.)
Lithium copper oxide (Li2CuO2) was synthesized by a simple solid state reaction involving a lithium source [lithium hydroxide monohydrate (LiOH·H2O, Kanto Chemical Co., Inc.)] and a high purity copper source [copper oxide (CuO, Kanto Chemical Co., Inc.)] at 850 °C for 24 h under O2 gas flow. The resultant powder was then stored in a dry room with a dew point of less than −40 °C. The same procedure was used for the cation, M (M = Co, Ni, Fe, Ti, and Mn), -substituted oxides. Mono oxides (Kanto Chemical Co., Inc.) for M = Co, Ni, Mn and metal powders (Kanto Chemical Co., Inc.) for M = Fe and Ti were used as cation sources. The crystallographic structure of the resultant powder was characterized with an X-ray diffractometer (Rigaku, RINT2500) using CuKα radiation under a constant power of 30 kV and 100 mA. The formation phases of the synthesized powder were identified using the database of the International Center for Diffraction Data (ICCD). The particle morphology was observed using a scanning electron microscope (SEM, SU1510, Hitachi High-Technologies Corp.).
(e) (d) (c) (b) (a) Li CuO(PDF:01-079-1975) 2
2.2. Electrochemical measurements of Li2Cu1 − xMxO2 (M = Co, Ni, Fe, Mn, and Ti) Electrochemical measurements were carried out using a twoelectrode coin-type test cell (type 2320). An electrode material mixture was prepared with Cu-based oxide (70 wt.%), Ketjen Black EC600JD (25 wt.%, Lion Co., Ltd.), and polytetrafluoroethylene (5 wt.%, Daikin Co., Ltd.). The mixture was then roll-pressed into a 0.5 mm thick sheet. This sheet was cut into a circle shape and used as a working electrode in a coin-type test cell. Lithium metal (0.3 mm thick) was used as a counter electrode. Microporous polypropylene film (Celgard), which was located between the two electrodes, was used as a separator. The electrolyte was 1 mol/L of LiPF6 in an equal volume of ethylene carbonate (EC) and dimethyl carbonate (DMC) solution (Kishida Chemical Co., Ltd.). The entire preparation and assembly process was carried out in a dry room with dew point of less than −40 °C. The tests on the coin-type cell were conducted in a constant temperature-controlled chamber maintained at 20 °C. The electrochemical measurement was carried out at a constant current density of 1 mA/cm2 in a voltage range of 2.0–4.0 V, with a rest period of 10 min between each charge and discharge step. 2.3. Stability evaluation of Li2Cu1 − xMxO2 (M = Co, Ni, Fe, Mn, and Ti) An ex-situ evaluation of the discharged/charged electrodes containing the Cu-based oxides was carried out to study the stability of the crystal structure when it underwent an electrochemical process. Test cells were performed under the same condition as the electrochemical evaluation (current density: 1 mA/cm2) but they were terminated under different conditions on the cut-off voltages or discharge/charge capacities. After the electrochemical evaluation, the cells were disassembled and the pellet electrode containing the Cu-based oxide was removed from the cell and washed several times with dimethylcarbonate solution (DMC, Kishida Chemical Co., Ltd.) to eliminate the lithium salt particles in the electrode. The pellet electrodes were dried and stored in a vacuum condition prior to the XRD measurements. 3. Results and discussion Fig. 1 shows XRD patterns of Li2Cu0.9M0.1O2 [M = Cu (unsubstituted), Co, Ni, Fe, Mn, and Ti]. Almost all the peaks for the cation-substituted Li2CuO2 corresponded to PDF data #01-079-1975 (space group: Immm) for Li2CuO2, although there were some new peaks or peak shifts in the patterns. No peaks indexed to the raw materials of monoxides or metal
15
20
25
2
30
35
40
2θ (deg) Fig. 1. XRD patterns of Li2Cu0.9M0.1O2 [M = (a) Cu (unsubstituted), (b) Co, (c) Ni, (d) Fe, (e) Ti, and (f) Mn].
were confirmed in the patterns. These results indicate that the cations can substitute the Cu sites in Li2CuO2. Morphology changes in the cation-substituted Li2CuO2 were analyzed by SEM observation. Fig. 2 shows SEM images of Li2Cu0.9Ni0.1O2 as a typical cation-substituted Cu-based oxide compared with Li2CuO2. The Nisubstituted Li2CuO2 particles are more homogeneous and have a smooth shape without a sharp edge compared with Li2CuO2. This result suggests that the crystallization temperature became lower as a result of the cation substitution, and the sintering of the particles can proceed easily. Fig. 3 shows the first charge and discharge curves of Li2Cu0.9M0.1O2 (M = Co, Ni, Fe, Ti, and Mn) electrodes compared with those of the unsubstituted Li2CuO2 electrodes. All the electrodes exhibited charge and discharge capacities of about 240 mAh/g and about 130 mAh/g, respectively. These values correspond to about 1.0 mol and about 0.6 mol equivalent of lithium in Li2Cu1 − xMxO2. Moreover, irreversible capacities of about 110 mAh/g were observed in all oxides, regardless of the substituent metals. As for Li2CuO2, this first discharge capacity agreed well with that of 130 mAh/g reported by Arai et al. [8]. Consequently, no clear improvement in the irreversible capacity could be achieved with a cation-substitution content, x = 0.1. However, it can also be seen that Ni- and Co-substituted Li2CuO2 electrodes have slightly larger discharge capacities than unsubstituted Li2CuO2. Fig. 4 shows the cycle properties of Li2Cu0.9M0.1O2 (M = Co, Ni, Fe, Ti, and Mn) electrodes compared with those of the unsubstituted Li2CuO2 electrodes. As shown in this figure, compared with the unsubstituted oxide, the Fe- and Mn-substituted oxides exhibited smaller discharge capacities and that of the Ti-substituted oxide was similar. On the other hand, the Ni- and Co-substituted oxides exhibited the larger discharge capacities than the unsubstituted oxide. In addition, the discharge capacities of the Ni-substituted oxide gradually increased as the cycle number increased. This unique behavior was not observed for any other cation-substituted oxide other than Ni. However, similar behavior has been observed in the cycle properties of Li2Cu0.8Ni0.2O2 as reported by Imanishi et al. [9], but they did not describe the reason for this behavior in detail. Fig. 5 shows the correlation between the Ni- and Co-substitution content, x, and the first charge and discharge capacities of Li2Cu1 − xMxO2
Please cite this article as: E. Setiawati, et al., Solid State Ionics (2013), http://dx.doi.org/10.1016/j.ssi.2013.11.043
E. Setiawati et al. / Solid State Ionics xxx (2013) xxx–xxx
(i)
3
(ii)
Fig. 2. SEM images of (i) Li2CuO2 and (ii) Li2Cu0.9Ni0.1O2.
(M = Co, Ni) electrodes. The charge capacities for both Ni- and Cosubstituted oxides deceased as the substitution content, x, increased. The Co-substituted oxides exhibited similar discharge capacities of about 140 mAh/g for the tested substitution condition and the x = 0.7 oxide exhibited almost no irreversible capacity. On the other hand, the Ni-substituted oxide exhibited the largest discharge capacity of about 160 mAh/g at x = 0.3. The x = 0.9 oxide exhibited the smallest irreversible capacity during the substation condition tested, although the oxide at this composition exhibited the smallest discharge capacity of about 100 mAh/g. As mentioned above, the Co and Ni substitution led to a significant decrease in the irreversible capacity. Such substitution of Cu sites by Co and Ni might suppress the transformation of the pristine Li2CuO2 structure, which was followed by a decrease in the irreversible capacity. As for Li2CuO2 and Ni-substituted oxides, the tendency on electrochemical properties agreed roughly with the previous reports [8–12]. However, the values of the charge/discharge capacities disagree somewhat with them. This may be due to the difference in the heat treatment condition. In this study, the heat treatment was carried out under O2 gas flow to enhance the cation-substitution reaction, whereas it was carried out in air [8,10] or in an Ar stream [9] in the previous reports. This difference in the calcination condition may have some effect on the oxidation state of Cu and Ni. Consequently, our results did not agree completely with those of the previous reports. Li2Cu0.1Ni0.9O2 and Li2Cu0.3Co0.7O2, which contained large amounts of rare metals such as Ni or Co, exhibited the smallest irreversible capacities. However, the properties of Li2Cu0.7Ni0.3O2 and Li2Cu0.9Co0.1O2, which contained less rare metals, were intensively investigated. These two oxides had characteristic properties. The former oxide exhibited the largest discharge capacity, and the latter exhibited the largest charge capacity. Fig. 6 shows the cycle properties of batteries with
4
Discharge capacity (mAh/g)
Voltage (V)
200
(f) (c)
3.5
(b)
(d) (a) (e) 3
(b) (f) 2.5
(e) (d)
2
Li2Cu0.7Ni0.3O2 and Li2Cu0.9Co0.1O2 cathode materials. Li2Cu0.9Co0.1O2 exhibited a stable capacity of about 130 mAh/g. On the other hand, the discharge capacity of Li2Cu0.7Ni0.3O2 gradually increased from 160 to 210 mAh/g as shown in Fig. 6(ii). A similar result was seen with Li2Cu0.9Ni0.1O2 as shown in Fig. 4(c). Of the oxides examined, the Nisubstituted Li2CuO2 showed this gradual increase in discharge capacity. The structural changes in the Cu-based oxides during the charge and discharge process were examined to understand the effect of substituting the Cu sites of Li2CuO2 by Ni and Co. Fig. 7 shows the XRD patterns of assynthesized Cu-based oxide powder and ex-situ XRD patterns of the electrodes containing the oxides after half charging (charge capacity: 125 mAh/g), full charging (cut-off voltage: 4.0 V), and full discharging (cut-off voltage: 2.0 V) at the first cycle. The unsubstituted Li2CuO2 structure transformed into LiCuO2 via an intermediate Li1.5CuO2 structure with a trace of Li2CuO2 during charging and then Li1.5CuO2 and Li2CuO2 coexist after discharging as shown in Fig. 7(i). The traces of Li2CuO2 that exist after full charging and full discharging are thought to be unreacted parts in the pristine material. After the subsequent cycle, the cycle seemed to be done between Li1.5CuO2 (discharging state) and Li1CuO2 (charging state) from a calculation of the discharge and charge capacities. These structural changes are consistent with those reported by Arai. The changes between the Li1.5CuO2 structure and the Li1CuO2 structure were also seen in the XRD patterns of the Co-substituted oxide. Therefore, these results suggest that the reaction mechanism for the Cosubstituted oxide would proceed in an almost similar manner to that for the unsubstituted oxide. As a result of the above-mentioned behavior, the unsubstituted and Co-substituted oxides exhibited almost the same irreversible capacities as shown in Fig. 3. However, there are differences between the unsubstituted and Co-substituted oxide as for the discharge products. After full discharging, the unsubstituted Li2CuO2 contained a mixed phase of Li2CuO2 and Li1.5CuO2, while the Li2Cu0.9Co0.1O2 contained only the Li1.5CuO2 phase. These results indicated that the
(c) (a)
150
100
0 0
50
100
150
200
250
300
Capacity (mAh/g) Fig. 3. First charge and discharge profiles for Li2Cu0.9M0.1O2 electrodes [M = (a) Cu (unsubstituted), (b) Co, (c) Ni, (d) Fe, (e) Ti, and (f) Mn].
: (a) Cu : (b) Co : (c) Ni
50
10
20
30
40
: (d) Fe : (e) Ti : (f) Mn 50
60
Cycle number Fig. 4. Cycle properties of Li2Cu0.9M0.1O2 electrodes [M = (a) Cu (unsubstituted), (b) Co, (c) Ni, (d) Fe, (e) Ti, and (f) Mn].
Please cite this article as: E. Setiawati, et al., Solid State Ionics (2013), http://dx.doi.org/10.1016/j.ssi.2013.11.043
4
E. Setiawati et al. / Solid State Ionics xxx (2013) xxx–xxx
CuKα
300
Capacity (mAh/g)
Charge
(i) Li CuO
: (i) Ni : (ii) Co
2
2
: Li2CuO2 (PDF #01-079-1975 ) : Li1.5CuO2 (PDF #01-079-1505 ) : Li1CuO2 (PDF #01-085-1926 )
250
200
(a) (b)
150
Discharge 100
0
0.2
0.4
(c) 0.6
0.8
1
(d)
x in Li2Cu1-xMxO2 Fig. 5. First charge and discharge capacities of Li2Cu1 − xMxO2 [M = (i) Ni, and (ii) Co] electrodes.
(ii) Li2Cu0.7Ni0.3O2
: Li2CuO2 (PDF #01-079-1975 ) : Li1.5CuO2 (PDF #01-079-1505 ) : Li1CuO2 (PDF #01-085-1926 ) : LiNiO2 (PDF #01-085-1966 )
(a) complete reversibility between Li2CuO2 and Li1.5CuO2 is not achieved by Co substitution. Moreover, Co substitution promotes the formation of layer-like structures Li1.5CuO2 and LiCuO2, and results in a slight increase in discharge capacity. On the other hand, structural changes between Li1.5CuO2 and Li1CuO2 structure were confirmed in the XRD patterns of Li2Cu0.7Ni0.3O2 shown in Fig. 7(ii). In addition, no peaks corresponding to the Li2CuO2 were found in the XRD patterns after full charging and full discharging as shown in Fig. 7(ii), (c) and (d). This result indicates that Ni-substitution promotes an electrochemical reaction between Li1.5CuO2 and LiCuO2 in a similar way with Co substitution. In addition to such changes, new peaks corresponding to LiNiO2 appeared at 2θ = 18.7 and 36.6°. To be precise, this phase should be noted as LixNiO2 because there were slight peak shifts in the XRD patterns seen in Fig. 7(ii), (b)–(d). These results indicate that the Ni-substituted oxide segregated two phases, LixCuO2 and LixNiO2. LixNiO2 is well-known as a cathode material for Li ion batteries and Li ions can be inserted/
(b) (c) (d)
(iii) Li2Cu0.9Co0.1O2
: Li2CuO2 (PDF #01-079-1975 ) : Li1.5CuO2 (PDF #01-079-1505 ) : Li1CuO2 (PDF #01-085-1926 )
(a) (b) 250
Discharge capacity (mAh/g)
(c) (d)
(i)
15
20
25
30
35
40
2θ θ (deg)
200
Fig. 7. XRD patterns of (a) as-synthesized Cu-based oxide powder and ex-situ XRD patterns of (b) half-charged, (c) charged, and (d) discharged electrodes containing Cubased oxides. (i) Li2CuO2 (unsubstituted), (ii) Li2Cu0.7Ni0.3O2 and (iii) Li2Cu0.9Co0.1O2.
150 extracted to/from this oxide. These results suggest that the formation of LixNiO2 is responsible for the characteristic behavior confirmed in the Ni-substituted oxides where there is a gradual increase in discharge capacities.
(ii)
100
4. Conclusions
5
10
15
20
25
Cycle number Fig. 6. Cycle properties of (i) Li2Cu0.7Ni0.3O2 and (ii) Li2Cu0.9Co0.1O2 electrodes.
Li2CuO2 exhibits two different electrochemical reactions; an irreversible one of Li2CuO2 to Li1.5CuO2 and a reversible one of Li1.5CuO2 to LiCuO2. Substituting Cu sites with transition metal ions was
Please cite this article as: E. Setiawati, et al., Solid State Ionics (2013), http://dx.doi.org/10.1016/j.ssi.2013.11.043
E. Setiawati et al. / Solid State Ionics xxx (2013) xxx–xxx
ineffective in improving the reversibility of Li2CuO2 and Li1.5CuO2, however Ni- and Co-substituted oxides showed a slight improvement in their reversible capacities. Both Ni and Co substitutions seemed to enhance the electrochemical reaction between Li1.5CuO2 and LiCuO2. Moreover, the formation of LixNiO2 in the Ni-substituted oxide caused a gradual increase in discharge capacity. Acknowledgments We thank Dr. Y. Sugiyama for his helpful guidance. We also thank Mr. H. Ota and Mr. T. Yamada for their help with the experiments.
5
References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12]
L. Lu, X. Han, J. Li, J. Hua, M. Ouyang, J. Power Sources 226 (2013) 272. C. Ahn, C. Li, H. Peng, J. Power Sources 196 (2011) 10369. M.S. Whittingham, Chem. Rev. 104 (2004) 4271. J.W. Fergus, J. Power Sources 195 (2010) 939. A.R. Wizansky, P.E. Rauch, F.J. Disalvo, J. Solid State Chem. 81 (1989) 203. K. Imai, M. Koike, H. Takei, H. Sawa, D. Shiomi, M. Kinoshita Nozawa, J. Phys. Soc. Jpn. 61 (1819) (1992). R. Berger, J. Less-Common Met. 169 (1991) 33. H. Arai, S. Okada, Y. Sakurai, J. Yamaki, Solid State Ionics 106 (1998) 45. N. Imanishi, K. Shizuka, T. Ikenishi, T. Matsumura, A. Hirano, Y. Takeda, Solid State Ionics 177 (2006) 1341. Y. Arachi, Y. Nakata, K. Hinoshita, T. Setsu, J. Power Sources 196 (2011) 6939. Y. Arachi, T. Setsu, T. Ide, K. Hinoshita, Y. Nakata, J. Power Sources 225 (2012) 611. C.T. Love, M.D. Johannes, A.M. Stux, K.E. Swider-Lyons, ECS Trans. 16 (29) (2009) 27.
Please cite this article as: E. Setiawati, et al., Solid State Ionics (2013), http://dx.doi.org/10.1016/j.ssi.2013.11.043