Abstract

Spherical precursor with narrow size distribution and high tap density has been successfully synthesized by a continuous hydroxide coprecipitation, and is then prepared by mixing the precursor with 6% excess followed by calcinations. The tap density of the obtained powder is as high as 2.61 g  . The powders are characterized by X-ray diffraction (XRD), X-ray photoelectron spectroscopy (XPS), scanning electron microscope (SEM), particle size distribution (PSD), and charge/discharge cycling. The XRD studies show that the prepared has a well-ordered layered structure without any impurity phases. Good packing properties of spherical secondary particles (about 12 μm) consisted of a large number of tiny-thin plate-shape primary particles (less than 1 μm), which can be identified from the SEM observations. In the voltage range of 3.0–4.3 V and 2.5–4.6 V, delivers the initial discharge capacity of approximately 175 and 214 mAh g−1 at a current density of 32 mA g−1, and the capacity retention after 50 cycles reaches 98.8% and 90.2%, respectively. Besides, it displays good high-temperature characteristics and excellent rate capability.

1. Introduction

During the past decade, lithium ion batteries have been extensively investigated and widely used; they are not only required to enable the moderately charge/discharge rates applications like mobile phone and portable computer but also to meet an increasing need for new applications such as electric vehicles, which need power sources with both high energy and high power density. Layered LiNi0.5Mn0.5O2 is of great interest as a promising cathode material for lithium secondary batteries because of its higher theoretical capacity (280 mAh g−1) and better structural stability [15]. However, some problems, such as uneasy preparation of stoichiometric phases [6], low tapping density [7], and poor rate capability [8], have to be overcome before it is massively applied in the lithium ion battery industry.

Recently, the effects of cobalt doping on the structure and electrochemical behavior of LiNi0.5Mn0.5O2 had been reported by Li et al. [9], and the results showed that cobalt doping for LiNi0.5Mn0.5O2 can easily form stoichiometric Li[Ni0.5Mn0.5−x Cox]O2 compounds, which possess good electronic conductivity, and thus owning good rate capability. In the series of LiNi0.5Mn0.5−x CoxO2, Li[Ni0.5Mn0.3Co0.2]O2 can be considered as one of the most promising cathode materials for the application of lithium ion battery, because this composition compromises between the increase of the discharge capacity due to the Co3+ and the increase of the thermal stability due to the Mn4+ ions. Nevertheless, the characteristics of the Li[Ni0.5Mn0.3Co0.2]O2 strongly depend on the preparation route. Liu et al. [10] stated that the Li[Ni0.5Mn0.3Co0.2]O2 prepared by traditionally mixed hydroxide method in flowing O2 showed homogeneous phase structure; the lattice parameters of the powder were a = 2.908 Å, c = 14.250 Å, and ratio of that represents the degree of cation mixing was 1.50. It delivered the initial discharge capacity of 150 mAh g−1 at a current density of 0.2 mA cm−2 between 2.75 and 4.2 V, but it showed a poor cycling behavior. Li et al. [11] have reported that Li[Ni0.5Mn0.3Co0.2]O2 synthesized by solid state reaction in O2 had an agglomerated morphology; the lattice parameters of the powder were a ≈ 2.87 Å, c ≈ 14.26 Å, with ratio of . It exhibited the initial discharge capacity of 172 mAh g−1, and the capacity retention after 25 cycles was less than 85% at a current density of 40 mA g−1 (0.4 mA cm−2) between 3.0 and 4.6 V. Above researches [10, 11] had proved that traditionally mixed hydroxide method and solid state reaction method were unsuitable for preparation of Li[Ni0.5Mn0.3Co0.2]O2 with high electrochemical performance and high tap density.

To obtain an ideal Li[Ni0.5Mn0.3Co0.2]O2 cathode material with high discharge capacity, excellent cycling performance and high volume energy density, the morphology and tap density of the material have to be controlled precisely during the preparation process. In general, high tap density particles could be obtained by increasing crystallinity and grain size of crystals, but this will lead to loss of specific discharge capacity. Another way to increase tap density of the material without loss of discharge capacity is to fabricate firstly the precursor of uniform-sized spherical particles by coprecipitation, then to obtain high performance Li[Ni0.5Mn0.3Co0.2]O2. In this paper, a continuous hydroxide coprecipitation method has been developed to prepare a spherical Li[Ni0.5Mn0.3Co0.2]O2 in a self-designed device. During the preparation process, two continuous stirred-tank reactors (CSTR) with same structure are used to control availability the morphology of the precursor. The one is used for coprecipitation reaction, and another is used for ageing process, which will prolong an average residence time of reaction particle in the vessel to get regular and round spherical precursor. The structural properties, morphologies, particle size distributions of the precursor [Ni0.5Mn0.3Co0.2](OH)2, and final product Li[Ni0.5Mn0.3Co0.2]O2 were discussed, and the electrochemical characteristics of the Li[Ni0.5Mn0.3Co0.2]O2 were also investigated in detail.

2. Experimental

2.1. Preparation of [Ni0.5Mn0.3Co0.2](OH)2 Precursor

Spherical [Ni0.5Mn0.3Co0.2](OH)2 was synthesized by a continuous hydroxide coprecipitation under N2 atmosphere. The details of the coprecipitation apparatus are shown in Figure 1. Initially, the CSTR-1 was filled with distilled water corresponding to 20 vol.% of the reactor, and the pH was then adjusted to 11.0 with proper amount of 3.6 M NH4OH and 3.6 M NaOH to get an initial solution. The solution was stirred at 800 rpm, and the temperature of the solution was maintained at 50°C by circulating hot water through the jacket of the reactor. The 1.8 M aqueous solution of NiSO4·6H2O, MnSO4·H2O, and CoSO4·7H2O corresponding to a molar composition of Ni : Mn : Co = 5 : 3 : 2 was introduced continuously into the CSTR-1 by a peristaltic pump. At the same time, a 3.6 M aqueous solution of NaOH and desired amount of complexant 3.6 M NH4OH were separately fed into the reactor. The total feed flow rate was adjusted to assure an average residence time of 10 h in CSTR-1. The liquid product that overflowed out from the CSTR-1 was feed into CSTR-2 for ageing step, which was kept at the same reaction conditions as CSTR-1. Finally, the products that collected continuously from the base of CSTR-2 were filtered, washed, and dried at 110°C for 12 h to obtain [Ni0.5Mn0.3Co0.2](OH)2 precursor.

2.2. Preparation of Li[Ni0.5Mn0.3Co0.2]O2

The [Ni0.5Mn0.3Co0.2](OH)2 precursor was mixed thoroughly with 6% excess of a stoichiometric amount of Li2CO3 (molar ratio of 1 : 1.06) to compensate the Li loss on the calcining process. The mixture was preheated at 500°C for 5 h and then calcined at 820°C for 12 h in air to obtain the final product Li[Ni0.5Mn0.3Co0.2]O2.

2.3. Characterization

The chemical composition of the resulting powder was analyzed by atomic absorption spectroscopy (Vario 6 Analytik Jena AG, Jena, Germany). The tap densities of [Ni0.5Mn0.3Co0.2](OH)2 and Li[Ni0.5Mn0.3Co0.2]O2 were determined by Powder Integrative Characteristic Tester (BT-1000, Bettersize Instruments Ltd, China), the vibration amplitude is 3 mm, the vibration frequency is 250 time min−1, and the repeatability precision is less than 1%. The particle size and particle size distribution were measured by Mastersizer-2000 (Malvern Instruments Ltd, England). The phase identification of the sample was performed with a diffractometer (D/Max-3C, Rigaku, Japan) using Cu Kα radiation (λ = 1.54178 Å) and a graphite monochromator at 36 kV, 20 mA. The scanning rate was 8° min−1, and the scanning range of diffraction angle ( ) was . X-ray photoelectron spectroscopy (XPS, PHI-5800) measurements were conducted to determine the sample. Al Kα (1486 eV) radiation was the primary excitation source, and the energy scale was adjusted based on the carbon peak in the C1s spectra at 284.6 eV. The morphology of the samples were observed using scanning electron microscopy (JSM-5600LV, JEOL, Japan).

The electrochemical tests of Li[Ni0.5Mn0.3Co0.2]O2 were carried out using coin cells assembled in an argon-filled glove box. In all cells, the cathode was consisted of a mixture of active material (80 wt.%), acetylene black (10 wt.%), graphite (5 wt.%), and polyvinylidene fluoride (5 wt.%) as a binder agent, lithium was served as counter and reference electrodes, a Celgard 2400 was used as separator, and the electrolyte was a 1 M LiPF6 solution in ethylene carbonate (EC)-dimethyl carbonate (DMC) (1 : 1 in volume). Charge/discharge measurements were carried out in Neware battery test system (BTS-XWJ-6.44S-00052, Newell, China).

3. Results and Discussion

Figure 2(a) displays the XRD pattern of the [Ni0.5Mn0.3Co0.2](OH)2 precursor. It can be observed that the XRD pattern of the [Ni0.5Mn0.3Co0.2](OH)2 is almost consistent with the typical fingerprint of Ni(OH)2 structure [12]. All diffraction lines are indexed to a hexagonal structure with a space group of p m1. The absence of impurity phases indicates that Ni, Co, and Mn would be homogeneously distributed within [Ni0.5Mn0.3Co0.2](OH)2 particle. The chemical composition of the [Ni0.5Mn0.3Co0.2](OH)2 precursor was analyzed by atomic absorption spectroscopy. The results identify that the element ratio of Ni : Mn : Co is 0.501 : 0.297 : 0.202, the element ratio of Ni : Mn : Co is nearly close to the design value of 5 : 3 : 2. Figure 2(b) illustrates the XRD pattern of the final product Li[Ni0.5Mn0.3Co0.2]O2. It reveals that the as-synthesized powder has the typical structure of a hexagonal α-NaFeO2 type with a space group of R m (no. 166). The diffraction peaks are quite narrow, indicating high crystallinity, and no impurity diffraction peak is observed. The splits in the (006)/(102) and (108)/(110) doublets indicate the formation of a highly ordered α-NaFeO2-type layered structure [13]. Lattice parameters, c/a value, value of the Li[Ni0.5Mn0.3Co0.2]O2 are tabulated in Table 1. The calculated lattice parameters of the powder are a = 2.870 Å, c = 14.256 Å, and V = 101.690 Å3. These data are similar to those observed by other researchers listed in Table 1. The value of c/a = 4.967 reveals that the layered structure is formed. In the layered structure, Li+, transition metal ions and oxygen ion are situated in Figures 3(a), 3(b), and 6(c), respectively. Since the ionic radii of Li+ (0.76 Å) and Ni2+ (0.69 Å) ions are similar, a partial disordering in Figures 3(a) and 3(b) is expected, and it is called “cation mixing” [14, 15]. It has been known that the cation mixing will deteriorate the electrochemical performance of the layered oxide materials. The integrated intensity ratio of (R) is sensitive to the cation mixing and usually taken as a measure of the amount of the cation mixing in the series of LiNiO2 families [16, 17]. Higher value of is desirable for lower amount of the cation mixing. R < 1.2 is an indication of undesirable cation mixing [18]. It can be seen from Table 1 that the value of R = 1.779 for the as-prepared Li[Ni0.5Mn0.3Co0.2]O2, which indicates the low amount of cation mixing in its layered structure.

Li[Ni0.5Mn0.3Co0.2]O2 compound was characterized by XPS in order to analyze the chemical composition of the spheres. The XPS emission spectra of Li[Ni0.5Mn0.3Co0.2]O2 and corresponding magnified Mn 2p3/2, Co 2p3/2 and Ni 2p3/2 are presented in Figure 3. In the spectrum of analysis of whole elements as shown in Figure 3(a), the peaks for Li, Mn, Co, Ni, and O originated from Li[Ni0.5Mn0.3Co0.2]O2 are observed. Besides, the peak for C can also be found in the spectrum, which probably came from CO2 that attached to the surface of the sample. As illustrated in Figure 3(b), the binding energy of Mn 2p3/2 electron in the Li[Ni0.5Mn0.3Co0.2]O2 is 641.7 eV, which is the same as that of the Li2MnO3. These results suggest that the valence of Mn is tetravalent, well consistent with those reported [19]. It can be seen from Figure 3(c) that the Co 2p3/2 has a characteristic peak with a binding energy of 779.6 eV and can be indexed to the Co3+ (the published binding energy of a trivalent Co 2p3/2 electron is 779.3–779.9 eV [20]). As shown in Figure 3(d), the Ni 2p3/2 peak is observed at 855.2 eV. Because the standard binding energies of Ni2+ and Ni3+ are 853.8 and 857.3 eV [21], respectively, it is expected that the valence number of Ni in Li[Ni0.5Mn0.3Co0.2]O2 is a mixture of 2+ and 3+. It is well known that the existence of mixed-valence cations (Ni2+/Ni3+) significantly contributes to the inherent electronic conductivity of electrodes during charge and discharge [22]. Therefore, good electrochemical performance would be expected for the as-prepared Li[Ni0.5Mn0.3Co0.2]O2.

Scanning electron micrographs of the [Ni0.5Mn0.3Co0.2](OH)2 precursor with different magnifications are presented in Figure 4. The secondary particle of precursor has an average size of about 12 μm with homogeneous spherical morphology, and it is composed of numerous primary particles with small size. The primary particle is a laminated flake minicrystal, which integrates closely into secondary particles during the precipitation process. Figure 5 shows the particle size distribution of [Ni0.5Mn0.3Co0.2](OH)2 precursor. It can be seen that there is a narrow and reasonable particle size distribution with the average grain size of 11.55 μm, which is consistent with the SEM observations. The tap density of the obtained precursor is up to 2.12 g cm−3, which can be attributed to the homogeneous distributions of spherical particles with good packing properties. It is well known that the particle morphology is a very important factor on the tap density of the powder. The powder composed of spherical particles has higher tap-density than the power composed of irregular particles. The reason may be as follows. In general, there are much less agglomeration and “bridge formation” of the particles within the powder composed of spherical particles, which result in less vacancy among the particles and excellent fluidity of the powder. During tapping, the small spherical particles with excellent fluidity can easily move and occupy the vacancies among the larger particles, which leads to a small quantity of space within the powder after long period tapping [23]. Cho et al. [24] reported that the shape and size of secondary particle of the final active material could, to a great extent, depend on the morphology and size of precursor. Therefore, it is expected that the uniform-sized [Ni0.5Mn0.3Co0.2](OH)2 with spherical morphology will result in a high tap density Li[Ni0.5Mn0.3Co0.2]O2.

SEM images of the as-prepared Li[Ni0.5Mn0.3Co0.2]O2 at different magnifications are illustrated in Figure 6. It can be seen that the secondary particle of the prepared Li[Ni0.5Mn0.3Co0.2]O2 has a similar size as those of the precursor [Ni0.5Mn0.3Co0.2](OH)2 even after the precursor has been recrystallized with lithium salt during high temperature calcination. Each of the spherical particles is made up of numerous primary grains as shown in Figure 6(c). However, the shape of the primary particles for Li[Ni0.5Mn0.3Co0.2]O2 has been changed to tiny-thin plate shape with an average diameter less than 1 μm, which is much different from that for the precursor. Figure 7 shows the particle size distribution of Li[Ni0.5Mn0.3Co0.2]O2. It can be observed that the average particle size is 12.25 μm, which is a little larger than that of the precursor. This is probably attributed to addition of the lithium source and expansion of the volume during the calcination process. The tap density of the prepared Li[Ni0.5Mn0.3Co0.2]O2 is as high as 2.61 g cm−3, which is close to that of commercialized LiCoO2 (2.70 g cm−3) [23, 25].

The specific capacity and capacity retention of the as-prepared Li[Ni0.5Mn0.3Co0.2]O2 were measured by a coin-type cell using Li[Ni0.5Mn0.3Co0.2]O2 as active material. Initial charge and discharge profiles of Li/Li[Ni0.5Mn0.3Co0.2]O2 cell under different charge/discharge conditions are shown in Figure 8, and the corresponding cycle performance is displayed in Figure 9. Figure 8(a) shows the cell voltage plotted versus specific gravimetric capacity for the initial charge/discharge cycle operated at a current density of 32 mA g−1 (0.2 C), in the voltage range of 3.0–4.3 V at 25°C. The prepared Li[Ni0.5Mn0.3Co0.2]O2 exhibits the initial discharge capacity of about 175 mAh g−1 and the capacity retention is as high as 98.8% after 50 charge/discharge cycling. Even though the charge/discharge current increases to 1 C rate, the Li[Ni0.5Mn0.3Co0.2]O2 still delivers an initial discharge capacity of 166 mAh g−1 (as shown in Figure 9), and the capacity retention still maintains 96.3% after 50 cycles. Li[Ni0.5Mn0.3Co0.2]O2 is only partially de-intercalated and intercalated when cells are charged and discharged in the voltage range of 3.0–4.3 V. Higher discharge capacities can be obtained when the voltage range is broadened. Figure 8(b) shows the initial charge/discharge cycle profile operated at 0.2 C between 2.5 and 4.6 V at 25°C. Initial charge and discharge capacity of 244 and 214 mAh g−1 are obtained, and the capacity of 193 mAh g−1 is retained at the end of 50 charge/discharge cycles with the capacity retention of 90.2%. Jouanneau et al. [26] recently reported that larger particles having higher tap-density would show less reactive at highly oxidized state in respect to thermal stability. Therefore, it is significantly critical to concern the high-temperature characteristics of the prepared Li[Ni0.5Mn0.3Co0.2]O2. Figure 8(c) presents the first charge/discharge cycle profile at 0.2 C between 3.0 and 4.3 V at 55°C. For elevated temperature test, Li/Li[Ni0.5Mn0.3Co0.2]O2 cell shows the same electrochemical behavior with the cell at room temperatures, and it delivers higher initial discharge capacities of 180 mAh g−1 but with similar cycling performance as compared to room temperature; 96.7% of an initial capacity is maintained after 50 cycles (as shown in Figure 9). This notable performance of the spherical Li[Ni0.5Mn0.3Co0.2]O2 might be due to the homogeneous distribution of transition metal ions, which originates from the preparation procedure applied to the [Ni0.5Mn0.3Co0.2](OH)2 precursor in order to maintain its layered structure. That is, the mixing of the three metal salt solutions before continuous coprecipitation results in a good atomic level distribution of transition metal ions in the prepared precursor of [Ni0.5Co0.2Mn0.3](OH)2. In addition, the layered structure of [Ni0.5Mn0.3Co0.2](OH)2 allows for ready formation of a well-ordered, layered structure in Li[Ni0.5Mn0.3Co0.2]O2. The initial discharge capacity and capacity retention of the spherical Li[Ni0.5Mn0.3Co0.2]O2 prepared in the continuous co-precipitator are better than the previous results of LiNi0.5Mn0.5−x CoxO2 ( ) materials prepared by other preparation methods [111, 27, 28]. Therefore, the preparation of spherical Li[Ni0.5Mn0.3Co0.2]O2 by continuous coprecipitation method could be a promising route to produce a cathode material for advanced lithium ion batteries.

It is well known that the rate capability can be affected strongly by surface texture and secondary particle size of cathode material. In order to evaluate the rate capability of the prepared Li[Ni0.5Mn0.3Co0.2]O2, the cell was charged galvanostatically with a 0.2 C (32 mA g−1) rate before each discharge testing, then discharged at different C rates from 0.2 to 10 C rates stepwise. Typical discharge curves of Li/Li[Ni0.5Mn0.3Co0.2]O2 cell in the voltage range of 3.0–4.3 V and 2.5–4.6 V at various current densities are shown in Figure 10, and their corresponding cycling performance are shown in Figure 11. From the capacity retention percentage based on 0.2 C (175 mAh g−1) between 3.0 and 4.3 V as shown in Figure 10(a), the Li[Ni0.5Mn0.3Co0.2]O2 delivers a discharge capacity of 171 mAh g−1 at 0.5 C (the capacity retention rate is about 97.7% of that of 0.2 C), 166 mAh g−1 at 1 C (the capacity retention rate is about 94.5% of that of 0.2 C), 153 mAh g−1 at 2 C (the capacity retention rate is about 88% of that of 0.2 C), and 135 mAh g−1 at 5 C (the capacity retention rate is about 78.5% of that of 0.2 C). Even at 10 C (1600 mA g−1), the capacity of Li[Ni0.5Mn0.3Co0.2]O2 is still as high as 116 mAh g−1 and the capacity retention rate is about 66.1% of that of 0.2 C. While the voltage range is broadened up to 2.5–4.6 V (as shown in Figure 10(b)), the Li[Ni0.5Mn0.3Co0.2]O2 still retains a high percentage of its discharge capacity at high currents during cycling process, the capacity retention percentage at 0.5 C, 1 C, 2 C, 5 C, and 10 C based on 0.2 C capacity (214 mAh g−1) are 95.7%, 92.5%, 85.4%, 74.5%, and 63.1%, respectively. The enhanced discharge capacity at accelerated rates clearly demonstrates the advantages of the spherical secondary particles. Compared with the irregular particles reported in literatures [3, 6, 9], the spherical particles with close-packed primary grains in this work will likely improve interparticle lithium ion movement, thereby enhancing the rate capability of the material.

4. Conclusions

Li[Ni0.5Mn0.3Co0.2]O2 prepared by the continuous hydroxide coprecipitation shows a homogeneous spherical morphology with well-ordered layered structure and has high tap density of 2.61 g cm−3, which is close to that of commercialized LiCoO2. The Li[Ni0.5Mn0.3Co0.2]O2 displays high initial discharge capacity of 175 mAh g−1 (3–4.3 V, 0.2 C, 25°C), 166 mAh g−1 (3–4.3 V, 1 C, 25°C), and 214 mAh g−1 (2.5–4.6 V, 0.2 C, 25°C), as well as good cycling performance. In addition, it exhibits good high-temperature characteristics and excellent rate capability. The good electrochemical performance of Li[Ni0.5Mn0.3Co0.2]O2 can be attributed to its structure integrity and the existence of mixed-valance cations (Ni2+/Ni3+). Therefore, it is concluded that the spherical Li[Ni0.5Mn0.3Co0.2]O2 powder prepared by the continuous coprecipitation method is a promising cathode material for advanced lithium ion batteries.

Acknowledgments

This paper was financially supported by the National Natural Science Foundation of China under project no. 20871101, scientific and technological plan project of Ministry of Science and Technology no. 2009GJD20021, Joint Fund of Natural Science of Hunan Province and Xiangtan City under project no. 09BG005, Industrial Project of Colleges and Universities of Hunan Province under project no. 10CY005, Project of Condition Research of Hunan Province under project no. 2010TC2004, and Colleges and Universities in Hunan Province plans to graduate research and innovation under project no. CX2009B133.