Full Text Article

Introduction

Humans pursue goals endlessly, and efforts to achieve them are expressed as energy conversion in any form. The representative ability of mankind is that mankind knows how to use tools, and mankind is making efforts to use tools as the most effective way to achieve its goals. Naturally, energy conversion is also necessary for the operation of the tool. Energy is neither created nor dissipated, just convertible. Efficient tool operation is realized through the highest efficiency energy conversion, and the energy associated with the highest conversion efficiency is electrical energy. Therefore, since the most efficient tool is electronic devices, electronic devices are indispensable to mankind for efficient and convenient life. It is generally overlooked or misunderstood that an electronic device without a battery, whether direct or indirect, may not exist. A battery is a device that converts various types of energy into electrical energy using chemical reactions, radiation, temperature differences, light activity, power, etc.

Because it is easy to design and manufacture lithium secondary batteries to best suit properties such as energy density, capacity, power, safety, life cycle, price, etc. depending on the unique demands of various electronic devices, they can be applied as a power source for a wide variety of electronic devices that require a wide range of energy density and power density.

Chemical formulas of commercially representative cathode materials, such as a layered crystal structure lithium transition metal oxide, a spinel-type crystal structure lithium transition metal oxide, and an olivine-type crystal structure lithium transition metal phosphate, can be abbreviated as LiMO₂ (M = Ni, Mn, Co, etc.), LiM^'₂O₄ (M' = Mn, Ni, Co, etc.), LiM^"PO₄ (M" = Fe, Mn, Ni, Co, etc.), respectively. Here, all of the M, M', and M" are 3d transition metals. This is because 3d transition metals have higher electrode potential than 4d and 5d transition metals, are relatively lightweight and small in size, which are advantageous in terms of battery capacity per unit weight and per unit volume [1].

Layered crystal structure LiNi_1-xNi_1-x-yMn_xCo_yO₂ (0.1 ≤ x + y ≤ 0.5) may exhibit a theoretical and also available capacity of up to 280 mAh/g as well as sustainable battery performances [1,2]. To obtain such LiNi_1-xNi_1-x-yMn_xCo_yO₂ , two significant challenges need to be addressed.

The first challenge is to ensure homogeneity in the arrangement of Ni, Mn, and Co within LiNi_1-xNi_1-x-yMn_xCo_yO₂ . A liquid reaction producing method can be easily and homogeneously mixed Ni, Mn, and Co in atomic units. As a solid reaction producing method, Ni, Mn, and Co cannot be homogeneously mixed in atomic units, and as a vapor phase reaction producing method, Ni, Mn, and Co can be homogeneously mixed in atomic units, but it is impossible to produce a powder-type cathode active material, and a film-type cathode active material can be produced. During charging and discharging of a lithium secondary battery made of LiNi_1-xNi_1-x-yMn_xCo_yO₂ as a cathode active material, Ni, Mn, and Co are oxidized and reduced in different sequences. When the arrangement of Ni, Mn, and Co within LiNi_1-xNi_1-x-yMn_xCo_yO₂ is heterogeneous, it can decrease the available capacity and shorten the overall battery lifespan. To overcome this issue, many efforts have been attempted, such as developments of particles achieving a concentration-gradient Ni, Mn, and Co as well as a hybridized shape of Ni-rich core and Mn-rich shell [3,4]. Regrettably, a cathode active material with a concentration-gradient Ni, Mn, and Co has not registered a noticeable market share.

The second challenge is the selective oxidation number control of nickel. Only when the average oxidation number of Ni, Mn, and Co of LiNi_1-x-yMn_xCo_yO₂ is +3 or slightly higher than +3, LiNi_1-x-yMn_xCo_yO₂ may have a layered crystal structure and maintain charge neutrality. Cobalt ion prefers the electron configuration of Co³⁺, while nickel and manganese ion prefers the electron configuration of Ni²⁺ and Mn⁴⁺ to Ni³⁺ and Mn³⁺ [1,5]. In the case of LiNi_1-x-yMn_xCo_yO₂ with high nickel ratio, single-phase layered crystal structure LiNi_1-x-yMn_xCo_yO₂ containing Ni²⁺ cannot be generated. Thus, three-dimensional crystal structure Li_1-zNi_1-x-y+zMn_xCo_yO₂ in which Li⁺ and Ni²⁺ are mixed in a non-stoichiometric ratio in each other, not in their own position, and/or impurities such as Li_1+zNi_1-x-y-zMn_xCo_yO₂ are inevitably mixed [1,6,7]. Therefore, only selective oxidation from Ni²⁺ to Ni³⁺ are required without altering the oxidation number of cobalt and manganese ions. Although it is known that one of the fundamental disadvantages of LiNi_1-x-yMn_xCo_yO₂ is an inappropriate but stable oxidation state of Ni²⁺, many indirect efforts have been made to solve this disadvantage, but unfortunately, there is no direct effort to solve this disadvantage [8-12]. It's important to note that the layered crystal structure LiNi_1-x-yMn_xCo_yO₂ is composed of a stacked combination of Ni_1-x-yMn_xCo_yO₂ layers and LiO₂ layers.

In this study, the solubility control depending temperature of coprecipitation conditions for homogeneous arrangement of Ni, Mn, and Co and the process conditions for only selective oxidation from Ni²⁺ to Ni³⁺ without altering the oxidation number of cobalt and manganese ions were established at the same time.

Experimental

A fabrication of a layered crystal structure Li_1-xNi_{1- x-y}Mn_xCo_yO₂ (0.1 ≤ x + y ≤ 0.5) with homogeneous arrangement of Ni³⁺, Mn⁴⁺, and C³⁺ may be summarized into three ways.

Homogeneous Arrangement of Ni, Mn, and Co in the Ni_1-x-yMn_xCo_y Coprecipitation Granules Through a Rapid Hydroxide Coprecipitation Method in 580^o C Aqueous Solution of Nickel Sulfate, Manganese Sulfate, and Cobalt Sulfate

The solubility of various Li salts and transition metal salts was evaluated according to temperature. For the production of Li_1-xNi_1-x-yMn_xCo_yO₂ (0.1 ≤ x + y ≤ 0.5) with homogeneous arrangement of Ni, Mn, and Co using a liquid reaction method, the evaluated solubilities were relatively compared to determine Ni salt, Mn salt, and Co salt having the same solubility at a specific temperature. These results are shown in Figure 1 and 2.

As shown in Figure 2, nickel sulfate, manganese sulfate, cobalt sulfate, and iron sulfate have the same solubility in 58^o C water. However, lithium sulfate and aluminium sulfate have a slightly lower solubility than nickel sulfate, manganese sulfate, cobalt sulfate, and iron sulfate in 58^o C water. The small difference in solubility between lithium sulfate and nickel/manganese/cobalt sulfates in 58^o C water indicates that aqueous ionization solution solgel synthesis method cannot be applied to the preparation method of single-phase layered crystal structure Li_1-xNi_1-x-yMn_xCo_yO₂ . The same solubility of nickel sulfate, manganese sulfate, and cobalt sulfate in 58^o C water represents that a precursor for producing Li_1-xNi_1-x-yMn_xCo_yO₂ with homogeneous arrangement of Ni, Mn, and Co can be produced through a rapid hydroxide coprecipitation method in 58^o C aqueous solution of nickel sulfate, manganese sulfate, and cobalt sulfate.

A mixed aqueous solution of potassium hydroxide (KOH), ammonium hydroxide (NH₄OH), and sodium hydroxide (NaOH) was selected as alkaline aqueous solutions for rapid hydroxide coprecipitation. Mixing molar ratio of potassium hydroxide (KOH), ammonium hydroxide (NH₄OH), and sodium hydroxide (NaOH) depends on the molar ratio of Ni, Mn, and Co in Li_1-xNi_1-x-yMn_xCo_yO₂ , as well as a fabrication amount. Here, it should be considered that the following three risk factors must be overcome. Since sodium ions have a smaller ionic radius than potassium ions, they may remain even after several filtrations and washings in the process of recovering secondary coprecipitation granules (result of collection, occlusion, and adsorption of primary coprecipitation particles). Because the remained sodium ions in secondary coprecipitation granules evidently remain in Li_1-xNi_1-x-yMn_xCo_yO₂ as cathode active material, they adversely affect lithium ion battery performances. Therefore, it is recommended to minimize the content of sodium hydroxide. Potassium hydroxide aqueous solution has the disadvantage of being highly toxic. In addition, ammonium hydroxide helps to uniformize particle size, collect primary particles, and spheroidize granules, but when its amount is excessive, it has a negative effect.

Selective Oxidation of Ni₂₊ up to Ni₃₊ without Changing the States of Co₃₊ and Mn₄₊ in Secondary Coprecipitation Granules

According to the characteristics of nickel ion and cobalt ion in the layered crystal structure Li_1-xNi_1-x-yMn_xCo_yO₂, both nickel ion and cobalt ion are located in the octahedral sites surrounded by six oxygen ions, have a low spin electron configuration, as well as an electronpairing energy (P) and an energy difference between t_2g and e_g orbital energy levels noted as 10Dq or Δ_oct is in a 6Dq < P < 10Dq relationship for nickel ion, and in a P < 6Dq relationship for cobalt ion. The crystal field stabilization energy is -18Dq + P for Co²⁺, 24Dq + 2P for Co³⁺, and -20Dq + 2P for Co⁴⁺. Therefore, a cobalt ion prefers the most stable electron configuration of Co³⁺. By the way, a nickel ion has the crystal field stabilization energy of 12Dq for Ni²⁺, 18Dq + P for Ni³⁺, and -24Dq + 2P for Ni⁴⁺, so it prefers the most stable electron configuration of Ni²⁺ [1]. A manganese ion prefers the most stable electron configuration of Mn⁴⁺.

In order that the average oxidation number of Ni, Mn, and Co of LiNi_1-x-yMn_xCo_yO₂ (0.1 ≤ x + y ≤ 0.5) is +3 or slightly higher than +3, the oxidation number of Ni in the LiNi_1-x-yMn_xCo_yO₂ should be +3.

Oxidants capable of increasing the oxidation state of transition metal ions such as Ni, Mn, and Co include potassium permanganate (KMnO₄ ), sodium perborate (NaBO₃ ), sodium perchlorate (NaClO₄ ), sodium chlorate (NaClO₃ ), sodium chlorite (NaClO₂ ), and sodium hypochlorite (NaClO). However, none of these oxidants selectively oxidized Ni²⁺ to Ni³⁺ without changing the states of Co³⁺ and Mn⁴⁺. These oxidants oxidized Co³⁺ to Co⁴⁺ and/or Ni²⁺ even further to Ni⁴⁺. Therefore, in order to control their oxidation power, various oxidizing agent-reducing agent hybrid composites were prepared by adding agent lower than their oxidation power with a different amount and then evaluated. As a result, a mixed aqueous solution of NaClO and hydrogen peroxide (H₂O₂ ) could selectively oxidize Ni²⁺ to Ni³⁺ without changing the states of Co³⁺ and Mn⁴⁺ of the secondary coprecipitation granules. The optimal mixing ratio and each concentration of the NaClO and H₂O₂ aqueous solutions vary depending on the composition mole ratio of Ni, Mn, and Co, as well as the amount of the secondary coprecipitation granules. However, if the concentration of the mixed aqueous solution of NaClO and H₂O₂ is higher than 0.05M, dissolution of the secondary coprecipitation granules may occur.

Fabrication of Li_1-xNi_1-x-yMn_xCo_yO₂ with Homogeneous Arrangement of Ni³⁺, Mn⁴⁺, and Co³⁺

Figure 3 shows the fabrication process diagram of Li_1-xNi_1-x-yMn_xCo_yO₂ (0.1 ≤ x + y ≤ 0.5) with homogeneous arrangement of Ni³⁺, Mn^4+, and Co³⁺. During the mixing of the secondary coprecipitation granules (Ni_1-x-yMn_xCo_y coprecipitate) and Li₂CO₃ , sucrose is added as an additive to prevent the agglomeration of the Li_1-xNi_1-x-yMn_xCo_yO₂ powders. The mixed molar ratio of Li₂CO₃ , secondary coprecipitation granules, and sucrose is 0.525 : 1 : 0.1.

The prepared Li_1-xNi_1-x-yMn_xCo_yO₂ (0.1 ≤ x + y ≤ 0.5) is characterized by means of Xray diffraction (XRD) spectroscopy, inductively coupled plasma-mass spectroscopy (ICP-MS), X-ray absorption near edge structure (XANES) spectroscopy, extended X-ray absorption fine structure (EXAFS) spectroscopy, and Li magic angle spin nuclear magnetic resonance (Li MAS NMR) spectroscopy.

Electrochemical lithium intercalation and deintercalation experiments of the prepared Li_1-xNi_1-x-yMn_xCo_yO₂ (0.1 ≤ x + y ≤ 0.5) were realized by charge and discharge of coin type lithium secondary batteries using a mixture of the products (94% in weight) with SuperP/VGCF (3% in weight) and polyvinylidene fluoride (PVdF, 3% in weight) as cathode, lithium foil as anode and a 1 M LiPF₆ solution in ethylene carbonate (EC) and ethyl methyl carbonate (EMC) (1 : 2, volumetric ratio) as electrolyte.

Results and Discussion

Figure 4 illustrates the X-ray diffraction (XRD) pattern of commercialized LiNi_0.8Mn_0.1Co_0.1O₂ and Li_0.9Ni_0.8Mn_0.1Co_0.1O₂manufactured according to the change in the mixing ratio and concentration of NaClO and H₂O₂ aqueous solution under the conditions for keeping the states of Ni³⁺, Mn⁴⁺, and Co³⁺.

The intensity ratio of the (003)/(104) diffraction peaks, symbolized as I(003)/I(104), for commercialized LiNi_0.8Mn_0.1Co_0.1O₂ is lower than the cation mixing recognition criterion of 1.2, but I(003)/I(104) for all four Li_0.9Ni_0.8Mn_0.1Co_0.1O₂ is higher than 1.2. The results of I(003)/I(104) and [I(006)+I(102)]/I(101) Bragg peak intensity analysis, as well as peak splits in the (006)/(102) and (108)/(110) doublets, and show that all four Li_0.9Ni_0.8Mn_0.1Co_0.1O₂ do not contain a representative impurity with a space group P3112 structure as well as the cation mixing phenomenon do not occur in the all four Li_0.9Ni_0.8Mn_0.1Co_0.1O₂ [13-15]. Thus, the XRD analysis reveals that all four Li_0.9Ni_0.8Mn_0.1Co_0.1O₂ have a single-phase layered crystal structure with a space group R-3m despite changes in the mixing ratio and concentration. These results indirectly show the successful preparation of Li_0.9Ni_0.8Mn_0.1Co_0.1O₂ with Ni³⁺ , Mn⁴⁺, and Co³⁺.

Figure 5 illustrates the Ni X-ray absorption near edge structure (XANES) spectroscopic results of NiO, commercialized LiNi_1-x-yMn_xCo_yO₂ manufactured by the general coprecipitation-heat treatment method (CP-NMC), and Li_{1- x}Ni_1-x-yMn_xCo_yO₂ prepared by this study (SH-NMC). In addition, CP-NMC622 is an abbreviation for commercialized LiNi_0.6Mn_0.2Co_0.2O₂ produced by the general coprecipitation-heat treatment method. For consult, SH-NMC, SHNMF and SH-NMA stand for Li_1-xNi_1-x-yMn_xCo_yO₂ , Li_1-xNi_1-x-yMn_xFe_yO₂ and Li_1-xNi_1-x-yMn_xAl_yO₂ prepared by this study, respectively. According to Coulomb's law, as the positive and negative charges increase, the attraction between the positive and negative charges strengthens. Therefore, as the Ni oxidation number increases, the electron binding energy required to separate electrons from Ni ion increases. According to the results of the Ni XANES spectroscopy analysis in Figure 5, the Ni oxidation numbers of NiO, CP-NMC, and SH-NMC are +2, value between +2 and +3 (indicating the presence of both Ni²⁺ and Ni³⁺), and +3, respectively

Mn XANES spectroscopy results and Co XANES spectroscopy results of CPNMC and SH-NMC are summarized in Figs. 6 and 7, respectively. According to the results of Mn and Co XANES spectroscopy, the Mn and Co oxidation numbers of CPNMC are +4 and +3, respectively, as well as the Mn and Co oxidation numbers of SHNMC are also +4 and +3.

Moreover, the Ni, Mn, and Co XANES spectroscopy results of SH-NMC show that Ni³⁺, Mn⁴⁺, and Co³⁺ are all located in octahedral sites through their pre-edge absence and white line shapes, as well as there is no hybridization of d electron orbits and p electron orbits

Consequently, it is confirmed that layered crystal structure Li_1-xNi_1-x-yMn_xCo_yO₂ (0.1 ≤ x + y ≤ 0.5) with Ni³⁺ , Mn⁴⁺, and Co³⁺ may be prepared.

Figure 8 is the results of Fourier transform into r-space of k3 -weighted Ni K-edge extended X-ray absorption fine structure (EXAFS) spectroscopy information of NiO, CP-NMC, and SH-NMC. Figure 8 also represents the indirect support for Ni oxidation number of Li_1-xNi_1-x-yMn_xCo_yO₂ through the results of Ni-O bond length.

As the difference in oxidation number between cations and anions intensifies, the attraction between cations and anions becomes stronger, and the bond length between cations and anions becomes shorter. Thus, the fact that the Ni-O bond lengths of NiO, CP-NMC, and SH-NMC are gradually shortened in order means that the Ni oxidation numbers of NiO, CP-NMC, and SH-NMC are gradually increasing in order.

Ni, Mn, and Co have nuclear spin quantum numbers of 7/2, 5/2, and 3/2, respectively. Their nuclear proton spins have magnetic spin-spin interactions not only with neighboring proton spins but also with electron spins. Here, the difference between substances occurs. Ni³⁺ has six paired 3d electrons and one unpaired 3d electron, while Mn⁴⁺ has three unpaired 3d electrons. As Ni³⁺ and Mn₄₊ have unpaired electrons, both Ni³⁺ and Mn⁴⁺ exhibit paramagnetism. Therefore, magnetic spin-spin interaction can be performed. On the other hand, Co³⁺ and Ni²⁺ have diamagnetism because they have only six paired 3d electrons. The difference is shown in Figure 9 as single signal peak and multiple signal peaks, which are the Li magic angle spin nuclear magnetic resonance (Li MAS NMR) microstructure information.

As Li_1-xNi_1-x-yMn_xCo_yO₂ was prepared through a rapid hydroxide coprecipitation at the same solubility of nickel sulfate, manganese sulfate, and cobalt sulfate in 580 C aqueous solution as well as a selective oxidation process of Ni²⁺ to Ni³⁺ without changing the states of Co³⁺ and Mn⁴⁺, it is expected that Li_1-xNi_1-x-yMn_xCo_yO₂ was made of homogeneous arrangement of Ni₃₊, Mn₄₊, and Co₃₊. By the way, LiNi_1-x-yMn_xCo_yO₂ produced through general hydroxide coprecipitation at room temperature has a concentration gradient characteristic of each component due to the precipitation of each component in a different order depending on the difference in solubility of each component. Strictly, LiNi_1-x-yMn_xCo_yO₂ with a concentration gradient is a mixture, not a solid solution capable of controlling material properties. For this reason, paramagnetic components and dimagnetic components are distinguished. In this case, the diamagnetic Co³⁺, the diamagnetic Ni²⁺ , the paramagnetic Ni³⁺ and the paramagnetic Mn⁴⁺ are individually and locally concentrated, so the Li MAS NMR spectrum of LiNi_1-x-yMn_xCo_yO₂ produced through general hydroxide coprecipitation at room temperature might show a combination of broad single signal peak and sharp multiple signal peaks. Because the Li MAS NMR spectrum of Li_1-xNi_1-x-yMn_xCo_yO₂ prepared by this study does not show such the combination, it is considered that Li_1-xNi_1-x-yMn_xCo_yO₂ prepared by this study was made of homogeneous arrangement of Ni³⁺ , Mn⁴⁺, and Co³⁺.

Figure 10 illustrates charge and discharge curves of a coin cell lithium secondary battery manufactured using SH-NMC811 as a cathode active material at 0.2C rate and 25^o C. The theoretical discharge capacity of a lithium secondary battery manufactured with LiNi_0.8Mn_0.1Co_0.1O₂ is 275.47 mAh/g, and the actual discharge capacity of a lithium secondary battery manufactured with SHNMC811 is 261.84 mAh/g. This is 95.1% of the theoretical discharge capacity per weight. Considering the partially irreversible energy conversion that always occurs in the first charging and discharging cycle, the coulombic efficiency ((discharge capacity)/(charge capacity)) and energy efficiency ([(coulombic efficiency)x(average discharge potential)]/(average charge potential)) of the lithium secondary battery manufactured using SH-NMC811, evaluated from second charging and discharging cycle to 40 charging and discharging cycles, are 99.7% and 99.2%, respectively, which represent very high energy conversion efficiency and also very high battery capacity retention rates. The partially irreversible conversion caused by the formation of solid electrolyte interphase (SEI) films is a necessary evil that inevitably contributes to the stabilization, safety and capacity decrease of a lithium secondary battery [1].

Figure 11 shows charge and discharge curves of a coin cell lithium secondary battery manufactured using SHNMC622 as a cathode active material at 0.2C rate and 250 C. The theoretical discharge capacity of a lithium secondary battery manufactured with LiNi_0.6Mn_0.2Co_0.2O₂ is 276.48 mAh/g, and the actual discharge capacity of a lithium secondary battery manufactured with SH-NMC811 is 242.73 mAh/g. This is 87.8% of the theoretical discharge capacity per weight. The lithium secondary battery produced with SH-NMC622 also experiences typical irreversible conversion in the first charging and discharging cycle. From the second charging and discharging cycle onwards, it demonstrates good stability, with high energy conversion efficiency and a nice battery capacity maintenance rate.

Figure 12 illustrates charge and discharge curves of a coin cell lithium secondary battery manufactured using SH-NMC523 as a cathode active material at 0.2C rate and 250 C. The theoretical discharge capacity of a lithium secondary battery manufactured with LiNi_0.5Mn_0.2Co_0.3O₂ is 276.42 mAh/g, and the actual discharge capacity of a lithium secondary battery manufactured with SH-NMC523 is 238.69 mAh/g. This is 86.3% of the theoretical discharge capacity per weight. The lithium secondary battery produced with SH-NMC523 also experiences typical irreversible conversion in the first charging and discharging cycle. From second charging and discharging cycle to 40 charging and discharging cycles, it demonstrates a remarkable stability, a superior energy conversion efficiency, and an amazing battery capacity maintenance rate of 99.8%.

Figure 13 shows charge and discharge curves of a coin cell lithium secondary battery manufactured using SHNMC523 as a cathode active material at 0.2C rate and 600 C. The theoretical discharge capacity of a lithium secondary battery manufactured with LiNi_0.5Mn_0.2Co_0.3O₂ is 276.42 mAh/g, and the actual 600 C discharge capacity of a lithium secondary battery manufactured with SH-NMC523 is 244.22 mAh/g. This is 88.4% of the theoretical discharge capacity per weight. Unlike charging and discharging at 250 C, the lithium secondary battery manufactured with SHNMC523 exhibits very slight decrease battery capacity from the second charging and discharging cycle onwards at 600 C.

Conclusion

An application of a lithium secondary battery manufactured with LiNi_1-x-yMn_xCo_yO₂ is limited by intrinsic disadvantages of LiNi_1-x-yMn_xCo_yO₂ . To overcome the intrinsic disadvantages of LiNi_1-x-yMn_xCo_yO₂ such as inappropriate but stable oxidation state of Ni²⁺ as well as uneven arrangement of Ni, Mn, and Co within LiNi_1-x-yMn_xCo_yO₂ , the solubility control according to coprecipitation temperature for homogeneous arrangement of Ni, Mn, and Co, as well as the selective oxidation control from Ni²⁺ to Ni³⁺ without changing the states of Co³⁺ and Mn⁴⁺ are established. In this way, single-phase layered crystal structure Li_1-x-yMn_xCo_yO₂ (0.1 ≤ x + y ≤ 0.5) with homogeneous arrangement of Ni³⁺, Mn⁴⁺, and Co³⁺ has been prepared and characterized by means of XRD, XANES, EXAFS, and Li MAS NMR.

The battery performance of the Li_1-xNi_1-x-yMn_xCo_yO₂ (0.1 ≤ x + y ≤ 0.5) obtained by the developed preparation method shows a maximum available capacity of 262mAh/g and a capacity retention rate of 99.7% in 40 charging and discharging cycles. Additionally, the battery capacity of the lithium secondary battery made with Li_1-αNi_1-x-yMn_xCo_yO₂ increases as the ratio of Ni increases.

Acknowledgement

This work was supported by research fund of Chungnam National University.