Next Article in Journal / Special Issue
Surface-Modified Membrane as A Separator for Lithium-Ion Polymer Battery
Previous Article in Journal
Recent Advances in Enzymatic Fuel Cells: Experiments and Modeling
Previous Article in Special Issue
Electrolytes and Interphasial Chemistry in Li Ion Devices
 
 
Retraction published on 31 August 2011, see Energies 2011, 4(9), 1336.
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Optimization and Characterization of Lithium Ion Cathode Materials in the System (1 – x – y)LiNi0.8Co0.2O2 • xLi2MnO3 • yLiCoO2

Department of Mechanical Engineering, Colorado State University, Fort Collins, CO 80523, USA
*
Author to whom correspondence should be addressed.
Energies 2010, 3(4), 847-865; https://doi.org/10.3390/en3040847
Submission received: 17 February 2010 / Accepted: 1 April 2010 / Published: 21 April 2010
(This article belongs to the Special Issue Lithium-ion Batteries)

Abstract

:
There is ongoing effort to identify novel materials that have performance better than LiCoO2. The objective of this work is to explore materials in the system (1 – x – y) LiNi0.8Co0.2O2 • xLi2MnO3• yLiCoO2. A ternary composition diagram was used to identify sample points, and compositions for testing were initially chosen. Detailed characterization of the synthesized materials was done, including Rietveld Refinement of XRD data, XPS analysis for valence state of transition-metals, SEM for microstructure details, and TGA for thermal stability of the materials. Electrochemical performance showed that discharge capacities on the order of 230 mAh/g were obtained. Preliminary results showed that these materials exhibit good cycling capabilities thereby positioning these materials as promising for Li-ion battery applications.

1. Introduction

Pioneering effort by Goodenough in the field of Li-ion battery cathode materials has positioned Li-ion battery technology as a viable technology for portable electronic applications and, more recently, for transport applications, after the proof of concept demonstration (reversible insertion/extraction of lithium from layered TiS2) [1,2,3,4,5,6,7,8]. Accordingly, LiCoO2, LiNiO2, and LiMn2O4 materials were explored initially as cathode materials for Li-ion battery technology. Extensive R&D work has allowed successful implementation of LiCoO2 batteries by Sony in the early 1990’s [9]. The structural instability, high cost, and non-environmental friendliness of cobalt materials make exploration of LiNiO2 materials attractive for Li-ion batteries [10]. However, the cation ordering associated with LiNiO2 has prevented the full realization of this electrode material [11,12,13]. This led to the investigation of solid solution between LiNiO2 and LiCoO2 with other dopants such as Fe and Al, resulting in material systems such as LiNi1–yCoyO2, LixFeyNi1–yO2, and LiAl1/4Ni3/4O2. [14,15,16]. Two-dimensional layered materials based on Mn were explored by Peter Bruce et al. [17]. Additional improvements in the layered insertion materials were proposed by Ohzuku et al. for cathode materials with composition LiNi0.5Mn0.5O2 [18], in addition to the investigation of the system Lix(MnyNi1–y)2–xO2 by Bates [19]. Ohzuku reported novel LiMn1/3Ni1/3Co1/3O2 which was investigated further [20]. Significant contributions were made in improving the layered two-dimensional materials by the Ohzuku, Dahn, Kanno, Zhecheva and Stoyanova groups [21,22].
Stoyanova et al. reported the effect of manganese substitution for cobalt in solid solution (LiCoO2-LiMnO2), which led to the development of composite-type cathodes [23]. Solid solutions between LiMO2 (M = Cr) and Li2MnO3-layered materials with superior electrochemical performance were demonstrated by Dahn and other groups [24]. A new series of material with the general formula xLi2MO3 • (1 – x)LiMn0.5Ni0.5O2 (M = Ti, Mn, Zr) for x = 0 to 0.3 was reported [25]. Approaches to prevent the conversion of layered LiMnO2 to spinel were explored which led to subsequent improvements of the material [26]. Layered compositions representing solid solution between Li2MnO3, i.e., Li[Li1/3Mn2/3]O2 and LiMn0.5Ni0.5O2, were shown to exhibit high capacities [27]. The new series of cathodes representing a solid solution between layered Li2MnO3 • Li[Li1–xMn(2–x)/3Nix/3Cox/3]O2 was proposed by Manthiram et al. which resulted in a high capacity of 285 mAh/g with low irreversible capacity loss [28]. These studies show that new cathode materials based on structural investigation are possible among several sub-classes of layered materials such as LiCoO2, LiNi0.5Co0.5O2, LiNi1/3Co1/3Mn1/3O2, and combinations thereof. Insertion materials such as LiCo1–2xNixMnxO2 (which is a solid solution of LiCoO2 and LiNi0.5Mn0.5O2) and Li[NixLi1/3–2x/3Mn2/3–x/3]O2 (a solid solution of LiNi1/2Mn1/2O2 and Li2MnO3) can exhibit satisfactory performance. However, when charged to 4.8 V, due to simultaneous removal of Li+ and O2–, these materials resulted in high initial irreversible capacity losses, potentially leading to safety concerns [29,30].
Zhang et al. reported novel cathode materials by developing solid solutions between three layered compounds: LiNi1/2Mn1/2O2, LiCoO2, and Li2MnO3 (i.e., Li[Li1/3Mn2/3]O2). The materials belonging to this composition diagram were synthesized by conventional solid state reactions [31]. The points chosen were concentrated in the high LiNi1/2Mn1/2O2 content region of the composition diagram (top region of the diagram), and cells were cycled between 4.4 or 4.6 V, to 2.5 V. Samples tested at the higher charge voltage showed that increasing the Li[Li2/3Mn2/3]O2 content led to increased capacity and better cyclability, but also had increased irreversible capacity loss (ICL), whereas the LiCoO2 component primarily contributed to increased capacity accompanied by fading during cycling. LiNi0.8Co0.2O2 (~190 mAh/g) materials have been reported to have higher capacities than LiNi1/2Mn1/2O2 (~150 mAh/g) [18]. We envisaged that by replacing one of the end-members, LiNi1/2Mn1/2O2, with LiNi0.8Co0.2O2, better capacities than reported could be achieved. Therefore the objectives of our work are (a) build a ternary compositional diagram with LiNi0.8Co0.2O2 as one of the end members while retaining LiCoO2 and Li2MnO3 (i.e., Li[Li1/3Mn2/3]O2) at corners, and (b) choose select compositions that are spread over the entire region of the ternary diagram that also give good representation of the compositional space. Applying this approach would ensure that the select compositions integrated both Li2MnO3 (i.e., Li[Li1/3Mn2/3]O2) and LiCoO2 at high content levels and a “sweet spot” composition could be found that has better performance than reported. The material synthesis conditions were kept the same as that of Zhang et al. to retain the less-complex solid-state synthesis and minimize the effect of varying synthesis procedures.

2. Results and Discussion

2.1. Structural studies

The ideal replacement of the LiNi1/2Mn1/2O2 end member in the ternary diagram would be LiNiO2 as an end member. However, we found that the synthesis conditions employed (i.e., 975 °C, 4 hours, liquid nitrogen quench) have resulted in NiO formation only, and accordingly we have chosen LiNi0.8Co0.2O2 as a corner of the compositional diagram (Figure 1). It is known that introduction of small amounts of Co stabilizes the LiNiO2 structure, leading to high specific capacities [16]. Choosing this composition has now provided the advantage that a common synthesis condition be applicable to all the compositions of the ternary diagram (see Table 1 for compositions of materials chosen). This is indeed the case that the XRD showed the synthesized compounds are single-phase in nature (Figure 2).
The powder X-ray diffraction of these eight compositions is given in Figure 2. X-ray diffraction patterns showed that all the compositions crystallize in a R-3m space group and belong to the two-dimensional layered Li[Ni, Co]O2 class of materials. In comparison to JCPDF standard, X-ray diffraction patterns were indexed with lattice parameters a ~ 2.84 Å, and c ~ 14.1 Å [32]. The crystal structure details of all the compositions are presented in Table 1. Also it was shown that the intensity ratio of (0 0 3) and (1 0 4) peaks is an indication of degree of cation ordering of LiNiO2, and a value of <1.2 could lead to poor electrochemical performance [33]. We have calculated that the intensity ratios of the materials are 0.69, 1.22, 1.21, 0.90, 1.19, 1.29, 1.12, and 1.07 respectively for samples #1 to # 8. It is noteworthy that the best performing materials (such as sample 6) have values greater than 1.2. ICP-AES analysis of best compositions was performed, and the compositions calculated based on the ratio of the concentrations are in agreement with the synthesized compositions within the limits of error (Table 1).
The synthesized materials belong to layered compounds of α-NaFeO2 type which is considered of distorted type (FCC array distorted along the hexagonal c-axis). This distortion is reflected in the XRD patterns, and accordingly the peaks around 2θ = 37 split into (006, 102) and peaks close to 70 split into (108, 110) Miller indices. When the distortion is absent, the splitting of the peak will not appear and merge into a single peak (the structure is cubic with extreme case of cation mixing). Therefore the value of c/a ratio is evidence of the degree of ordering of the layered structure as well as the amount of transition metal in the inter-slab structure. For c/a ratios greater than 24 = 4.899, it is reasonable to expect that the layered structure is formed, overcoming the distortion of oxygen lattice. It is indeed the case in our samples, the calculated c/a ratio values are 4.933, 4.943, 4.942, 4.949, 4.943, 4.974, 4.989, and 4.937 respectively for samples #1 to #8 [34]. In order to determine the cationic distribution and thereby the crystal structure more accurately, detailed Rietveld refinement of the structural analysis was performed.

2.2. Rietveld structural refinement details

Two dimensional layered LiNiO2 materials are known to display cation disordering (i.e., a contrary to standard R-3m group where Li occupies the 3a position, Ni in the 3b, and O in the 6c position; the cation ordering results in some amount of Ni in 3a sites). This cation disorder (Li1–zNi1+zO2) was attributed to poor cycling performance of LiNiO2 material [33]. In order to understand the role of the different transition metals in this series, detailed Rietveld refinement was performed. Applying the oxidation state and ionic radius, a model for cationic distribution is derived. Considering the ionic radii of Ni3+ (0.56 Å), Ni2+ (0.70 Å) to that of Li+ (0.74 Å), Ni entering Li would be divalent in nature. Neutron diffraction experiments on LiNi1–yCoyO2 showed that Co ions do not exist at the 3a Li site (as Co3+) [35]. The Rietveld refinement of Li[NixLi1/3–2x/3Mn2/3–x/3]O2 by Dahn et al. has shown that a model with Mn atoms moving to the Li-layer has given a good fit initially for structural refinement for the XRD data [27]. However, neutron diffraction that was applied to identify the disordering of Ni and Mn atoms showed an excellent fit to the model when Ni atoms occupy positions in the Li layers. In addition, a model with Mn atoms moving into Li layers resulted in unphysical values for the site occupations. Accordingly, we decided to refine our data with the following models: (a) Mn atoms in Li layers and (b) Ni atoms in Li layers. The long-scan data showed that around 2θ = 20°, additional Bragg reflections were present. These were attributed to superlattice reflections due to ordering of Li and Mn atoms which reduces the symmetry from R-3m to C2/m, and these were accordingly taken into consideration during refinement. In both cases we have employed a two-phase approach to refine the XRD data (i.e., R-3m with C 2/m, to account for superlattice reflections). The best fit for our data is obtained for model b—i.e., Ni atoms present in Li layers. The Rietveld refinement results of both the models are presented in Figure 3. Rietveld results in conjunction with ionic radii showed a general formula of the cationic distribution for the materials: [Li1–zMz2+]3a [M3+1–z–2yM2+y+zMn4+y]3b[O2]6c, which would also account for charge compensation at the 3b site.
Table 2 gives crystal structural details as a result of the Rietveld refinement. The results of structural refinement are analyzed using CrystalMakerR software [36] to obtain the crystal structure diagrams (Figure 4), from which the relevant bond lengths and angles between metal-oxygen coordination were derived (Table 2) [36,29].
Table 2 showed the amount of transition metals at the 3a Li site for both the models (up to 1.7%). In addition, the data on interatomic distances also supported the “cationic distribution” model. The M-O bond lengths obtained are 2.012 Å, and 2.013 Å respectively in case of M = Mn and Ni respectively. The bond length of Co-O with low-spin Co3+ is 1.945 Å, and with high-spin Co2+ is 2.01 Å. Similarly the bond length of Ni-O with low spin Ni3+ is 1.96 Å, and high-spin Ni2+ is 2.01 Å [33]. Our bond-length results in comparison are suggestive that Ni2+ and or Co2+ are present at 3a site (Co3+ cannot be present per neutron diffraction studies) and Co3+, Mn4+ present at the 3b site. The redox changes associated with such transition-metals plays an important role in determining the electrochemical properties of the materials.

2.3. Microstructural studies

TGA was performed to evaluate the thermal stability of the material (Figure 5). Within the temperature range of study, the samples lose a small percentage of weight continuously (total 3%). Weight loss up to 300 °C (about 2.5 wt %) could be due to loss of water molecules [38]. The small weight loss (~0.5 wt %) up to 600 °C indicates there was so significant loss corresponding to the material and the material is stable. TGA of LiNiO2 showed that there was negligible weight loss up to 860°C followed by decomposition of LiNiO2 to Li1–xNi1+xO2 along with oxygen evolution [39]. TGA results of our samples showed that they are thermally stable within the testing range (our instrument had limited testing range). Figure 6 shows the SEM images of all the compositions synthesized. SEM showed that the oxide particles have controlled morphology, generally are submicron in size, and have homogeneous distribution. Also the particles are interconnected well, which would facilitate good transport properties.

2.4. XPS Studies

X-ray photoelectron spectroscopy (XPS) is a well established, non-destructive technique for evaluation of valence states of the metals/ions in solids and is extensively applied in the characterization of cathode materials [34]. We have applied this technique to determine the valence states of our cathode materials synthesized. The valence states of the transition metals provide an idea of the redox couples associated with the intercalation process of lithium.
The measurement was done on uncharged samples at room temperature to determine the valence states of transition metals such as Ni, Mn and Co by obtaining a full-scan spectrum (up to 1200 eV) which contains signatures of valence states of the elements. The binding energy of C1s peak at 248.8 eV was used as the reference [40]. In order to analyze and meet the objectives, the spectral data of each sample is split into regions of interest and such results are presented in Figure 7. For example, in Figure 7, #1 gives a spectral response corresponding to Ni, Co, Mn, and O atoms corresponding to the composition #1, by splitting into regions around 850, 780, 640, and 530 eV, respectively.
For sample #1, the binding energy of Ni, Ni2p3/2 showed an asymmetric peak in addition to the satellite peak (Figure 7a). The overall spectrum is deconvoluted into two Gaussian peaks, The binding energy of the main peak is at 854 eV, which is indicative of Ni in the 2+ valence state and compares with other Ni2+, containing compounds, such as NiO. The peak situated at higher energy (856 eV) corresponds to Ni3+, as observed in Ni3+ containing compounds such as LiNiO2. The shift indicates an increase in the positive interaction between the electrons of 2p3/2 and the nucleus increases resulting in higher energy needed to pull the electron from Ni3+ level (Ni3+ has fewer electrons). The satellite peak at 860 eV is attributed to the multiple splitting in the energy levels of the Ni-containing oxides such as NiO or LiNiO2. The XPS results showed that the predominant state of Ni in the synthesized material is 2+ [41].
Figure 7b shows the binding energy corresponding to Li 1s core levels in the region 52–56 eV. The fitting and deconvolution of Li (1s) spectrum gives the major peak binding energy at 54.5 eV, which is similar to lithium in an octahedral environment of oxygen atoms, such as in LiCoO2 [41]. Figure 7c shows the O 1s spectrum, which has a main peak around 529 eV and another peak at 533 eV that can be deconvoluted to three individual contributions [34,42]. The peak at 529 eV can be assigned to O2– ions, and the shoulder peak could represent a degree of oxidation more than O2- due to transition metal-oxygen covalent bonds. Accordingly, the binding energies observed at 529 eV, 531.3 eV, and 532.4 eV could be assigned as contributions from the Mn-O, Co-O, and Ni-O bonds, respectively, implying the influence of the covalent nature of their bonds [34]. The distribution of transition metal atoms contributed to the strong superlattice reflections as seen in the XRD patterns. The ordering of Li and transition metal atoms in the rock salt structure of the compound has a transition metal layer consisting of Ni2+, Mn4+, and Co3+, separated by Li+ layers, and the vertices of the hexagon occupied by Mn4+ and Co3+, with Ni2+ in the center. The ordering has resulted in reducing the symmetry from R-3m to C2/m.
Figure 7d shows the Co 2p core level spectra which show Co 2p3/2 peaks in the 778–782 eV range and Co 2p1/2 peaks in the 792–797 eV range. The spectrum also has weak satellite peaks. The Co 2p3/2 peak was fit into a single peak with the binding energy of 779.9 eV which matches with the binding energy reported for Co3+ in LiCoO2. The binding energy of Co is indicative of Co3+ in low spin configuration. The peak at 796 eV corresponding to Co 2p1/2 is assigned to high spin Co2+ ions [43].
The XPS spectrum in Mn 2p core levels is shown in Figure 7e, in which the 3/2 and 1/2 orbit doublet components are seen at 642 eV and 655 eV, respectively. The asymmetric nature of the spectrum suggested deconvolution of the spectrum to identify the major peak position which would indicate the predomoninant state of Mn. The deconvolution resulted in two binding energies for the best fit. The major peak centered at 642.5 eV, which in comparison with the binding energies of Mn, containing compounds such as MnO2 (major peak at 642.5 eV) and Mn2O3 (641.8 eV), strongly suggested the 4+ valence of Mn and accordingly, the value was assigned. The secondary peak is attributed to the energy level splitting of Mn 2p3/2 electronic levels [44]. Thus the predominant valence states of transition metals in our samples were determined to be Ni2+, Co3+, and Mn4+, respectively. It is to be noted that in the well-known oxide cathodes, like LiNiO2, Li (Ni, Co)O2, and Li(Mn, Co)O2, with the α-NaFeO2 structure, the metals Ni, Mn, and Co are only in a 3+ oxidation state [33].
XPS study was extended to other synthesized materials and the results are presented in Table 3. The binding energy peak positions of lithium, cobalt, nickel, and manganese are evidenced in the XPS spectra. The binding energy of lithium is at ~55 eV indicating that the lithium atoms are in an octahedral environment. The oxygen O (1s) core level was observed at 530.5 eV. The binding energy peaks of Co (2p3/2) and Co (2p1/2) are observed at ~779 eV and ~795 eV, respectively, attributing to the Co3+ bonding states of these materials in a possible octahedral arrangement. The Ni 2p XPS spectrum is indicative of the charge—transfer nature of an octahedral 2+ metal ion; however, the broad and splitting nature (2p3/2, 2p½) is suggestive of a valence state of Ni greater than +2. The binding energy separation of about 15 eV is in good agreement with previous results. The Ni 2p½ and 2p3/2 spectra are observed at ~875 eV and ~857 eV, respectively, with a binding energy separation of about ~17 eV. Most of the compounds showed Mn 2p½ to Mn 2p3/2 peak separation at ~11 eV. The summary of XPS results showed that the predominant oxidation states of Ni, Co, and Mn are +2, +3, and +4 respectively [45]. This result implies that the primary redox couples involved in the electrochemical process will be Ni2+/Ni4+ and Co3+/Co4+ as the Mn4+ ion does not oxidize in the voltage window used.

2.5. Electrochemical results

The electrochemical performances of all compositions in the voltage range 4.6 to 2 V are presented in Figure 8. The system showed fairly consistent charge/discharge curves throughout the composition range, and the discharge capacities varied between 140 and 230 mAh/g in the voltage range of 4.6 to 2 V. The discharge capacities of these compositions are presented in Table 1. The discharge profiles were nearly flat with a mid Voc of approximately 3.8 eV. Sample #6 showed a long gradual drop resulting in the highest capacity when discharged to 2V.

2.6. Discussion

In general, the materials showed discharge capacities in the 140–230 mAh/g range (Table 2). We have looked at the specific capacity of the materials by dividing the overall capacity plot into four regions: 4.6–3 V, 4.6–2.75 V, 4.6–2.5 V, and 4.6–2 V, to comment on the usefulness of these materials for applications. The results are presented in Table 2. In the 4.6–3 V region, the best capacity is obtained for sample # 6 (190 mAh/g). Also sample #2 has nearly the same capacity at the 3 V level which is often taken as the cutoff of a practical cycle. When the range is extended to 2.75 V, sample #6 provides >200 mAh/g capacity, with sample #2 closely (>190 mAh/g). At the 2 V range there are four materials that have capacities >200 mAh/g. It is noteworthy that sample #6 has the highest discharge capacity at every charge level with the highest reproduced results of 230 mAh/g. The discharge profiles of the compositions, as shown in Figure 9a, exhibited a nearly flat, but clearly sloping pattern. Also the preliminary cyclability tests as shown in Figure 9b exhibit good capacity retention for the best composition under testing conditions (composition # 6).
The redox processes associated in the electrochemical process are derived from the charge/discharge curves. During the charging process, Li+ extraction occurs during charging from 3.4 V to about 4.4 V with concomitant oxidation of Ni2+ and Co3+, but around 4.5 V involves electrochemical activation of Li2MnO3 resulting in irreversible lithium and oxygen removal in the form of Li2O and MnO2 (Li2MnO3 → MnO2 + Li2O). During the discharge, lithium insertion takes place at about 4.3 V, with a smooth sloping profile (with a mid Voc at about 3.8 V of the discharge curve) that could correspond to reducing Ni4+/Ni2+ electrochemical redox couple and Co4+ reducing to Co3+, and then into a layered MnO2 component below 3.5 V down to 2 V (with a Mn4+/Mn3+ redox couple). This has resulted in gradual insertion of Li into the layered structure and the extended discharge profile has a specific capacity of about 230 mAh/g.
Results showed that the best composition, #6, with composition Li1.2Mn0.4Ni0.16Co0.24O2, in the range 4.6–2.5 V, had a discharge capacity of 215 mAh/g compared to ~200 mAh/g for Li1.133Mn0.517Ni0.250Co0.100O2, which was the highest capacity in Zhang’s results [31]. The results showed the effect of integrating Li2MnO3 (i.e., Li[Li1/3Mn2/3]O2) and LiCoO2 into the structure as well as optimizing the regions of high manganese and cobalt content. This approach has helped to identify materials with high capacity. It is noteworthy that sample #6 corresponding to 60% Li2MnO3 content had an ICL of ~60 mAh/g (in comparison to projected higher ICL of Zhang’s results by extrapolation at the same Li2MnO3 content).
It is noteworthy to mention that sample #6 had the highest Mn content (0.4 vs. 0.27 for the next highest #7). Integrating 2-D layered Li2MnO3 into isostructural components has indeed resulted in improved performance. It is known that when charging Li2MnO3 to 4.5 V, only 20 mAh/g can be extracted [46], but on charging to 5.0 V a capacity of 383 mAh/g (83% of the material’s theoretical capacity) can be achieved. This increased capacity is due to the electrochemical extraction of Li and the chemical extraction of oxygen that occurs at potentials >4.5 V [25a]. Therefore, charging sample #6 beyond 4.5 V to 4.6 V has helped to get additional charge capacity. The charging to 4.6 V does not result in complete removal of oxygen from the structure, allowing some amount of lithium to be utilized during the intercalation process. This helps to maintain good cyclability of the material, but also results in some ICL. Efforts are in progress to look at compositions around the most promising compositions of this system with high Mn content and utilizing modified synthesis procedures that help to obtain a high capacity material that is cost-effective and has lower ICL.

3. Experimental Section

3.1. Synthesis

All the compositions of the system were synthesized by conventional solid state reactions. Accordingly, stoichiometric quantities of the required precursors (lithium acetate, manganese acetate, nickel acetate, and cobalt acetate) were thoroughly mixed using mortar and pestle, and the powders were subjected to an initial heating at 450 °C for 8 hours to convert the precursors to basic oxides. Then the powders were ground, pelletized, and subjected to a second heating at 975 °C for 4 hours before quenching in liquid nitrogen.

3.2. Characterization

The phase purity and crystal structure details were examined by Scintag X-ray diffraction using Cu-Kα radiation. For Rietveld refinement, data were collected at a scan rate of 0.5°/min, with a step size of 0.2° in 2θ, over a range of 10° to 90°. The data were refined using the Rietveld analysis program FullProf [47,48]. The morphology of the powder was examined using a JEOL JSM-840A SEM, a field emission system with the in-lens thermal field emission electron gun. Elemental analysis of the compositions was performed by the ICP-AES method (Jarrell-Ash dual view, high resolution IRIS Advantage inductively coupled plasma). Thermogravimetric analysis (TGA) was performed using a weight and temperature-calibrated TA Instrument 2160. The samples were heated in a platinum pan with a heating rate of 5° C/min in order to reach final temperature. X-ray Photoelectron Spectroscopy (XPS) experiments were performed on a Physical Electronics PHI 5800 ESCA system. This system has a monochromatic Al Kα X-ray source (hα= 1486.6 eV), hemispherical analyzer, and a multichannel detector. A low energy (30 eV) electron gun was used for the charge neutralization on the non-conducting samples. The binding energy for the samples was referenced to the C1s peak at 284.8 eV. XPSPEAK Version 4.1 peak fitting software was used to deconvolute individual peak components. Peaks were fitted using the Gaussian-Lorentizian peak profile and a Shirley background was used.

3.3. Electrochemical testing

The electrochemical performance of oxide cathodes was tested using a T-cell configuration. The active material was mixed with conducting carbon and PTFE in a 75:20:5 ratio before rolling into thin sheets. A thin sheet of lithium metal was used as the anode. 1M LiPF6 [EC: EMC: DMC 1:1:1 by volume] was used as the electrolyte. The T-cells were assembled in a VAC manufactured glove box under high purity Ar gas. The T-cells were tested using an Arbin cycler under galvanostatic mode under constant current (0.1 mA/cm2) between 4.6 V to 2 V.

4. Conclusions

We have synthesized a novel series of (1 – x – y)LiNi0.8Co0.2O2 • xLi2MnO3 • yLiCoO2 compounds from the ternary compositional diagram between isostructural layered LiNi0.8Co0.2O2, Li2MnO3, and LiCoO2 materials by well-established solid-state methods by selecting compositions that give a good representation of the composition space. Structural details were elucidated by applying Rietveld structural refinement which showed the occupation of Ni atoms in Li layers confirming the cation distribution in these types of materials. The microstructural details elucidated through SEM showed the particles are sub-micron size with interconnectivity between particles. TGA showed good thermal stability for these materials in the temperature range of study. XPS results showed that the predominant valence states of Ni, Co, and Mn are 2+, 3+, and 4+ respectively. Electrochemical testing of the samples showed the advantage of integrating the isostructural 2-D layered systems such as Li2MnO3 and LiCoO2 resulting in increased capacity, good cyclability and considerably lower ICL. Discharge profiles showed association of redox couples Ni4+/Ni2+ in conjunction with Co4+/3+ (up to ~3.5 V), followed by Mn4+/3+, up to 2V. Synthesized materials exhibit discharge capacities in the 140–230 mAh/g range, and the approach helped to identify the material with the highest capacity of 230 mAh/g corresponding to the composition of Li1.2Mn0.4Ni0.16Co0.24O2. Efforts are currently underway to optimize the performance of the material further.

Acknowledgements

We thank Allan Kirkpatrick, Head, Department of Mechanical Engineering, for his continued help, support and encouragement.

References and Notes

  1. Mizushima, K.; Jones, P.C.; Wiseman, P.J.; Goodenough, J.B. LixCoO2 (0 < x < 1): A new cathode material for batteries of high energy density. Mater. Res. Bull. 1980, 15, 783–789. [Google Scholar]
  2. Thomas, M.G.S.R.; David, W.I.D.; Groves, P.; Goodenough, J.B.; Groves, P. Synthesis and structural characterization of the normal spinel Li[Ni2]O4. Mater. Res. Bull. 1985, 20, 1137–1146. [Google Scholar] [CrossRef]
  3. Goodenough, J.B.; Hong, H.Y.P.; Kafalas, J.A. Fast Na+-ion transport in skeleton structures. Mater. Res. Bull. 1976, 11, 203–220. [Google Scholar] [CrossRef]
  4. Whittingham, M.S. Electrical energy storage and intercalation chemistry. Science 1976, 192, 1126–127. [Google Scholar] [CrossRef] [PubMed]
  5. Tarascon, J.M.; Armand, M. Issues and challenges facing rechargeable lithium batteries. Nature (London) 2001, 414, 359–3676. [Google Scholar] [CrossRef] [PubMed]
  6. Thackeray, M.M.; David, W.I.F.; Bruce, P.G.; Goodenough, J.B. Electrochemical extraction of lithium from LiMn2O4. Mater. Res. Bull. 1989, 19, 179–187. [Google Scholar] [CrossRef]
  7. Padhi, A.K.; Nanjundaswamy, K.S.; Masquelier, C.; Goodenough, J.B. Mapping of transition metal redox energies in phosphates with NASICON structure by lithium intercalation. J. Electrochem. Soc. 1997, 144, 2581–2586. [Google Scholar] [CrossRef]
  8. Padhi, A.K.; Nanjundaswamy, K.S.; Goodenough, J.B. Phospho-olivines as positive-electrode materials for rechargeable lithium batteries. J. Electrochem. Soc. 1997, 144, 1188–1194. [Google Scholar] [CrossRef]
  9. Nishi, Y. Lithium Ion Batteries; Wahihara, M., Yamamoto, O., Eds.; Wiley-WCH, Kodansha: Tokyo, Japan, 1988. [Google Scholar]
  10. Goodneough, J.B.; Kizushima, K.; Takada, T. Solid-solution oxides for storage-battery electrodes. Jpn. J. Appl. Phys. 1980, 19, 305–313. [Google Scholar] [CrossRef]
  11. Rougier, A.; Saadoune, I.; Gravereau, P.; Willmann, P.; Delmas, C. Effect of cobalt substitution on cationic distribution in LiNi1–yCoyO2 electrode materials. Solid State Ionics 1996, 90, 83–90. [Google Scholar] [CrossRef]
  12. Morales, J.; Perez-Vicente, C.; Tirado, J.L. Cation distribution and chemical deintercalation of Li1–xNi1+xO2. Mater. Res. Bull. 1990, 25, 623–630. [Google Scholar] [CrossRef]
  13. Dahn, J.R.; Sacken, U.; Michael, C.A. Structure and electrochemistry of Li1 ± yNiO2 and a new Li2NiO2 phase with the Ni (OH)2 structure. Solid State Ionics 1990, 44, 87–97. [Google Scholar] [CrossRef]
  14. Rougier, A.; Saadouane, I.; Gravereau, P.; Willmann, P.; Delmas, C. Effect of cobalt substitution on cationic distribution in LiNi1–yCoyO2 electrode materials. Solid State Ionics 1996, 90, 83–90. [Google Scholar] [CrossRef]
  15. Saardoune, I.; Delmas, C. On the LixNi0.8Co0.2O2 system. J. Solid State Chem. 1998, 136, 8–15. [Google Scholar] [CrossRef]
  16. Ohzuku, T.; Yanagawa, T.; Kouguchi, M.; Ueda, A. Innovative insertion material of LiAl1/4Ni3/4O2 (R-3m) for lithium-ion (shuttlecock) batteries. J. Power Sources 1997, 68, 131–134. [Google Scholar] [CrossRef]
  17. Armstrong, A.R.; Bruce, P.G. Synthesis of layered LiMnO2 as an electrode for rechargeable lithium batteries. Nature (London) 1996, 381, 499–500. [Google Scholar] [CrossRef]
  18. Ohzuku, T.; Makimura, Y. Layered lithium insertion material of LiNi0.5Mn0.5O2: A possible alternative to LiCoO2 for advanced Lithium-ion batteries. Chem. Lett. 2001, 30, 744–745. [Google Scholar] [CrossRef]
  19. Neudecker, B.; Zuhr, R.A.; Kwak, B.S.; Bates, J.B.; Robertson, J.D. Lithium manganese nickel oxides Lix(MnyNi1–y)2–xO2. J. Electrochem. Soc. 1998, 145, 4148–4159. [Google Scholar] [CrossRef]
  20. Ohzuku, T.; Makimura, Y. Layered lithium insertion material of LiCo1/3Ni1/3Mn1/3O2 for lithium-ion batteries. Chem. Lett. 2001, 30, 642–643. [Google Scholar] [CrossRef]
  21. Johnson, C.S.; Kim, J.S.; Kropf, J.; Kahaian, A.J.; Vaughey, J.T.; Fransson, L.M.L.; Edstrom, K.; Thackeray, M.M. Structural characterization of layered LixNi0.5Mn0.5O2 (0 < x < 2) oxide electrodes for Li batteries. Chem. Mater. 2003, 15, 2313–2322. [Google Scholar]
  22. Alcantara, R.; Lavela, P.; Tirado, J.L.; Zhecheva, E.; Stoyanova, R. Recent advances in the study of layered lithium transition metal oxides and their application as intercalation electrodes. J. Solid State Electrochem. 1999, 3, 121–134. [Google Scholar]
  23. Stoyanova, R.; Zhecheva, E.; Zarkovam, L. Effect of Mn-substitution for Co on the crystal structure and acid delithiation of LiMnyCo1–yO2 solid solutions. Solid State Ionics 1994, 73, 233–234. [Google Scholar] [CrossRef]
  24. Ammundsen, B.; Paulsen, J.M. Novel Lithium-ion cathode materials based on layered manganese oxides. Adv. Mater. 2001, 13, 943–956. [Google Scholar] [CrossRef]
  25. Kim, J.S.; Johnson, C.S.; Vaughey, J.T.; Thackeray, M.M.; Hackney, S.A.; Yoon, W.; Grey, C.P. Electrochemical and structural properties of xLi2M’O3 • (1 – x)LiMn0.5Ni0.5O2 electrodes for lithium batteries (M’ = Ti, Mn, Zr). Chem. Mater. 2004, 16, 1996–2006. [Google Scholar] [CrossRef]
  26. Shin, S.; Sun, Y.K.; Amine, K. Synthesis and electrochemical properties of Li[Li(1–2x)/3NixMn(2–x)/3]O2 as cathode materials for Lithium secondary batteries. J. Power Sources 2002, 112, 634–638. [Google Scholar] [CrossRef]
  27. Lu, Z.; Beaulieu, L.Y.; Donaberger, R.A.; Thomas, C.L.; Dahn, J.R. Synthesis, structure, and electrochemical behavior of Li[NixLi1/3–2x/3Mn2/3–x/3]O2. J. Electrochem. Soc. 2002, 149, A778–A781. [Google Scholar] [CrossRef]
  28. Wu, Y.; Manthiram, A. High capacity, surface-modified layered Li[Li(1–x)/3Mn(2–x)/3Nix/3Cox/3]O2 cathodes with low irreversible capacity loss. Electrochem. Solid-State Lett. 2006, 9, A221–A224. [Google Scholar] [CrossRef]
  29. Lu, Z.; Chen, Z.; Dahn, J.R. Lack of cation clustering in Li[NixLi1/3–2x/3Mn2/3–x/3]O2 (0 < x ≤ 1/2) and Li[CrxLi(1–x)/3Mn(2–2x)/3]O2 (0 < x < 1). Chem. Mater. 2003, 15, 3214–3220. [Google Scholar]
  30. Yabuuchi, N.; Ohzuku, T. Novel lithium insertion material of LiCo1/3Ni1/3Mn1/3O2 for advanced lithium-ion batteries. J Power Sources 2003, 119, 171–174. [Google Scholar] [CrossRef]
  31. Zhang, L.; Takada, K.; Ohta, N.; Fukuda, K.; Osada, M.; Wang, L.; Sasaki, T.; Watanabe, M. Layered (1 – x – y)LiNi1/2Mn1/2O2 • xLi[Li1/3Mn2/3]O2 • yLiCoO2 cathode materials. J. Electrochem. Soc. 2005, 152, A171–A178. [Google Scholar] [CrossRef]
  32. JCPDF Standard 44-0145(for LiCoO2) and 9-0063 (for LiNi1–xCoxO2). International Centre for Diffraction Data: Newtown Square, PA, USA, January 2010.
  33. Ohzuku, T.; Ueda, A.; Nagayama, M.; Iwakoshi, Y.; Komori, H. Comparative study of LiCoO2, LiNi1/2Co1/2O2 and LiNiO2 for 4 Volt secondary lithium cells. Electrochim. Acta 1993, 38, 1159. [Google Scholar] [CrossRef]
  34. Rougier, A.; Gravereau, P.; Delmas, C.J. Optimization of the composition of the Li1–zNi1+zO2 electrode materials: Structural, magnetic, and electrochemical studies. J. Electrochem. Soc. 1996, 143, 1168–1175. [Google Scholar] [CrossRef]
  35. Shaju, K.M.; Subba Rao, G.V.; Chowdari, B.V.R. Performance of layered Li(Ni1/3Co1/3Mn1/3]O2 as cathodes for Li batteries. Electrochim. Acta 2002, 48, 145–151. [Google Scholar] [CrossRef]
  36. Liu, H.; Zhang, Z.; Gong, Z.; Yang, Y. Structural, electrochemical and thermal properties of LiNi0.8–yTiyCo0.2O2 as cathode materials for lithium ion battery. Electrochimica Acta 2004, 49, 1151. [Google Scholar]
  37. CrystalMaker®: A Crystal and Molecular Structures Programs for Mac and Windows; CrystalMaker Software Ltd.: Oxford, UK, January 2010.
  38. Saadoune, I.; Delmas, C. LiNi1–yCoyO2 positive electrode materials: relationships between the structure, physical properties and electrochemical behavior. J. Mater. Chem. 1996, 6, 193–199. [Google Scholar] [CrossRef]
  39. Kim, W.S.; Chung, K.; Choi, Y.K.; Sung, Y.E. Synthesis and charge-discharge properties of LiNi1–x–y CoxMyO2 (M = Al, Ga) compounds. J. Power Sources 2003, 115, 101–109. [Google Scholar] [CrossRef]
  40. Wang, G.X.; Zhong, S.; Bradhurst, S.; Dou, S.X.; Liu, H.K. Synthesis and characterization of LiNiO2 compounds as cathodes for rechargeable batteries. J. Power Sources 1998, 76, 141–146. [Google Scholar] [CrossRef]
  41. Endo, E.; Yasuda, T.; Yamamura, K.; Kita, A.; Sekai, K. LiNiO2 electrode by chemical vapor deposition for higher voltage performance. J. Power Sources 2001, 93, 87–92. [Google Scholar] [CrossRef]
  42. Moses, A.M.; Flores, H.G.; Kim, J.G.; Langell, M.A. Surface properties of LiCoO2, LiNiO2 and LiNi1–xCoxO2. Appl. Surf. Sci. 2007, 253, 4782–4791. [Google Scholar] [CrossRef]
  43. Galakhov, V.R.; Kurmaev, E.Z.; Uhlenbrock, S.; Newmann, M.; Kellerman, D.G.; Gorshkov, V.S. Electronic structure of LiNiO2, LiFeO2 and LiCrO2: XPS and X-ray emission study. Solid State Commun. 1995, 95, 347–351. [Google Scholar] [CrossRef]
  44. Kalai Vani, V.; Hussain, O.M. Synthesis and characterization of electron beam evaporated LiCoO2 thin films. Ionics 2007, 13, 473–477. [Google Scholar] [CrossRef]
  45. Kosova, N.V.; Devyatkina, E.T.; Kaicheve, V.V. LiMn2O4 and LiCoO2 cathode materials obtained by mechanical activation. Russ. J. Electrochem. 2009, 45, 277–285. [Google Scholar] [CrossRef]
  46. Mansour, A.N. Characterization of LiNiO2 by XPS. Surf. Sci. 1994, 3, 279. [Google Scholar] [CrossRef]
  47. Johnson, C.S.; Kim, J.S.; Lefief, C.; Vaughey, J.T.; Thackeray, M.M. The significance of the Li2MnO3 component in 'composite' xLi2MnO3 • (1 – x)LiMn0.5Ni0.5O2 electrodes. Electrochem. Commun. 2004, 6, 1085–1091. [Google Scholar] [CrossRef]
  48. Wiles, D.B.; Young, R.A. A new computer program for Rietveld analysis of X-ray powder diffraction patterns. J. Applied Crystallogr. 1981, 14, 149–151. [Google Scholar] [CrossRef]
  49. Roisnel, T.; Rodriguez-Carjaval, J. Fullprof Manual; Institut Laue-Langevin: Grenoble, France, January 2010. [Google Scholar]
Figure 1. Compositional ternary diagram representing regions of materials studied.
Figure 1. Compositional ternary diagram representing regions of materials studied.
Energies 03 00847 g001
Figure 2. Powder XRD patterns of samples of the (1 – x – y)LiNi0.8Co0.2O2 • xLi2MnO3 • yLiCoO2 system of compositions, marked from #1 to # 8, whose compositions are given in Table 1.
Figure 2. Powder XRD patterns of samples of the (1 – x – y)LiNi0.8Co0.2O2 • xLi2MnO3 • yLiCoO2 system of compositions, marked from #1 to # 8, whose compositions are given in Table 1.
Energies 03 00847 g002
Figure 3. Rietveld refinement results of composition # 6 using different models.
Figure 3. Rietveld refinement results of composition # 6 using different models.
Energies 03 00847 g003
Figure 4. Crystal structures of composition #6 created with CrystalMaker software.
Figure 4. Crystal structures of composition #6 created with CrystalMaker software.
Energies 03 00847 g004
Figure 5. TGA of synthesized materials.
Figure 5. TGA of synthesized materials.
Energies 03 00847 g005
Figure 6. SEM images of synthesized materials. All images are taken at 5000 magnification and the scale bar represents 1 µm.
Figure 6. SEM images of synthesized materials. All images are taken at 5000 magnification and the scale bar represents 1 µm.
Energies 03 00847 g006
Figure 7. XPS spectra of all the compositions showing the electronic transitions of elements present. Measured and fitted transition-metal spectra of the # 6. (ᴼ) represents the measured data points, (closed symbols) individual curve, ( -   - ) Shirley background, and (-) summation curve. Peak assignments are discussed in the text.
Figure 7. XPS spectra of all the compositions showing the electronic transitions of elements present. Measured and fitted transition-metal spectra of the # 6. (ᴼ) represents the measured data points, (closed symbols) individual curve, ( -   - ) Shirley background, and (-) summation curve. Peak assignments are discussed in the text.
Energies 03 00847 g007
Figure 8. Electrochemical testing of all the compositions showing the first charge and discharge curves at 0.1 mA/cm2.
Figure 8. Electrochemical testing of all the compositions showing the first charge and discharge curves at 0.1 mA/cm2.
Energies 03 00847 g008
Figure 9. (a) Discharge voltage profiles of all the compositions. (b) preliminary extended life-cycle testing of the materials.
Figure 9. (a) Discharge voltage profiles of all the compositions. (b) preliminary extended life-cycle testing of the materials.
Energies 03 00847 g009
Table 1. Compositions, crystal structure, and electrochemical results of (1 – x – y) LiNi0.8Co0.2O2 • xLi[Li1/3Mn2/3]O2 • yLiCoO2.
Table 1. Compositions, crystal structure, and electrochemical results of (1 – x – y) LiNi0.8Co0.2O2 • xLi[Li1/3Mn2/3]O2 • yLiCoO2.
CompositionsSpace Groupa
Å
c
Å
Discharge capacity (mAh/g)
4.6–3 V4.6–2.75 V4.6–2.5 V4.6–2 V
#1Li1.033Mn0.067Ni0.64Co0.26O2R-3m2.883(2)14.222(3)118122128141
#2Li1.1Mn0.200Ni0.48Co0.22O22.869(5)14.182(1)186192198209
#3Li1.033Mn0.067Ni0.48Co0.42O2 *2.857(1)14.119(5)177184190216
#4Li1.133Mn0.267Ni0.32Co0.28O22.853(2)14.119(3)172177182191
#5Li1.067Mn0.133Ni0.32Co0.48O22.848(6)14.080(2)177183189199
#6Li1.2Mn0.4Ni0.16Co0.24O2 *2.842(3)14.136(2)190203215230
#7Li1.133Mn0.267Ni0.16Co0.44O22.85(1)14.220(4)166175185199
#8Li1.067Mn0.133Ni0.16Co0.64O22.856(2)14.101(1)164171179192
* Formula as per ICP-AES are Li1.02Mn0.06Ni0.51Co0.39O2 (sample #3) and Li1.18Mn0.39Ni0.15Co0.26O2 (sample #6).
Table 2. Rietveld refinement results of composition #6.
Table 2. Rietveld refinement results of composition #6.
Lattice parameter (Å)z(O)Li3aOccupancy of M in Li layerχ2Rwp
(%)
RB
(%)
Bond Length(Å)Bond Angle(°)
Model 1a = 2.842(2)
c = 14.136(7)
0.249(1)0.988z(Mn) 0.0151.157.9515.8Mn-O 2.012
Li-O 2.029
O-Mn-O 90.11
O-Li-O 88.96
Model 2a = 2.842(2)
c = 14.136(7)
0.247(3)0.982z(Ni) 0.0171.218.1012.6Ni-O 2.013
Li-O 2.027
O-Ni-O 89.79
O-Li-O 89.05
Table 3. Summary of XPS results of the compositions investigated.
Table 3. Summary of XPS results of the compositions investigated.
SampleMn2p1/2Mn2p3/2ΔMnNi2p1/2Ni2p3/2ΔNiCo2p1/2Co2p3/2ΔCo
#1-642.8-871.685417.6794.0778.915.1
#2-642.0-871.6854.816.8794.0779.614.4
#3-642.0-871.6854.816.8794.8779.615.2
#4653.2642.810.4870.8854.016.8794.0778.815.2
#5653.2642.011.2870.8854.016.8794.0778.815.2
#6652.4642.010.4870.1853.216.9794.0778.815.2
#8652.4641.211.2869.8853.316.5793.2778.015.2

Share and Cite

MDPI and ACS Style

Manivannan, V.; Chennabasappa, M.; Garrett, J. Optimization and Characterization of Lithium Ion Cathode Materials in the System (1 – x – y)LiNi0.8Co0.2O2 • xLi2MnO3 • yLiCoO2. Energies 2010, 3, 847-865. https://doi.org/10.3390/en3040847

AMA Style

Manivannan V, Chennabasappa M, Garrett J. Optimization and Characterization of Lithium Ion Cathode Materials in the System (1 – x – y)LiNi0.8Co0.2O2 • xLi2MnO3 • yLiCoO2. Energies. 2010; 3(4):847-865. https://doi.org/10.3390/en3040847

Chicago/Turabian Style

Manivannan, Venkatesan, Madhu Chennabasappa, and Joshua Garrett. 2010. "Optimization and Characterization of Lithium Ion Cathode Materials in the System (1 – x – y)LiNi0.8Co0.2O2 • xLi2MnO3 • yLiCoO2" Energies 3, no. 4: 847-865. https://doi.org/10.3390/en3040847

Article Metrics

Back to TopTop