Electrochemical performances of Li-rich layered-layered Li2MnO3-LiMnO2 solid solutions as cathode material for lithium-ion batteriesContents lists avai
journal homepage: http: / /www.elsevier .com/locate/ ja lcom
Electrochemical performances of Li-rich layered-layered Li2MnO3- LiMnO2 solid solutions as cathode material for lithium-ion batteries
Rakesh Saroha a, Amit Gupta b, Amrish K. Panwar a, *
a Department of Applied Physics, Delhi Technological University, New Delhi, 110042, India b Department of Mechanical Engineering, Indian Institute of Technology Delhi, New Delhi, 110016, India
a r t i c l e i n f o
Article history: Received 19 September 2016 Received in revised form 12 November 2016 Accepted 15 November 2016 Available online 15 November 2016
Keywords: Layered-layered nanocomposite Sol-gel Electrochemical performance
* Corresponding author. E-mail address: panwaramar@gmail.com (A.K. Pan
http://dx.doi.org/10.1016/j.jallcom.2016.11.199 0925-8388/© 2016 Elsevier B.V. All rights reserved.
a b s t r a c t
In this work a series of Li-rich layered-layered solid solution of Li2MnO3-LiMnO2 [Li[Li(1x)/3Mn(xþ2)/3]O2; x ¼ 0.0, 0.1, 0.3 and 0.5] nanocomposite structure were successfully synthesized using sol-gel technique. All the synthesized compositions exhibit the main characteristics peaks of m-Li2MnO3 and could be indexed to the C2/m space group except some weak diffraction peaks located around 20e30 which can be attributed to the superlattice structure originate due to the ordering of Mn ion into Li-Mn layers and are typically observed for Li-rich based materials. CV results show that pristine LMO possesses weak anodic peak around 4.7 V and no symmetric cathodic peak in the voltage window of 2.0e4.8 V. Among all the synthesized compositions, x ¼ 0.3 (LMO3) delivers highest specific discharge capacity and best rate and cycling performances at all values of current densities. The LMO3 composition delivers an initial discharge capacity of 177 ± 5 mAhg1 at a current density of 10 mA/g in the voltage range of 2.0e4.8 V and holds nearly 97% of the initial discharge capacity after 120 charge/discharge cycles at the same current density.
© 2016 Elsevier B.V. All rights reserved.
1. Introduction
Single layered-type transition metal oxide cathode materials such as ABO2/AB2O3, spinel-type cathodes such as AB2O4 and olivine-type cathodes materials such as ABXO4 (A ¼ Li, B¼ transition metal, X¼ P, Si, etc.) have been extensively studied in order to have high energy density electrode material. Apart from advantages possesses by these materials their disadvantages are also evident. For ABO2-type cathode materials such as LiCoO2 and LiNiO2, higher cost, toxic nature and poor thermal stability remain important points of concerns [1,2]. In AB2O4-type cathodes such as LiMn2O4, capacity fading due to the dissolution of manganese ions and Jahn-Teller distortion hindered their widespread applications [3,4]. Furthermore, poor electronic conductivity and sluggish lithium ion diffusion in olivine-type cathode materials such as LiFePO4 restrict their use in automotive industry [5e7]. Layered- layered solid solutions of type AB2O3-ABO2 have become attrac- tive cathode materials for lithium-ion batteries due to their high reversible capacity of 250 mAhg1 with lower cost, structural sta- bility and better safety as compared to the currently used
war).
commercial LiCoO2 cathode [8]. Layered oxide Li[Li1/3Mn2/3]O2 (commonly written as Li2MnO3) possess AB2O3-type monoclinic structure (C2/m space group) and high theoretical capacity up to 458mAhg1 [9e11]. However, it is electrochemically inactive in the typical voltage range 2.2e4.4 V vs. Liþ/Li as all Mn atoms exist inþ4 valence. Due to this, it is hard to oxidize beyond Mn4þ in an octa- hedrally coordinated environment and therefore, no lithium can be easily removed from the Li2MnO3 phase [2,3,8,9,11]. Furthermore, it requires a high cut-off voltage (~4.8 V vs. Liþ/Li) during the initial charging process to activate Li2MnO3 by removing Li2O from the electrode active material according to the following reaction [2,9,10,12,13]:
Li2MnO3 / xLi2O þ (1 x)MnO2 (1)
Li2MnO3 also shows poor cycling performance which is due to the structural transition from layered structure to the spinel structure during charging/discharging process [14]. Therefore, rigorous research efforts are on-going to find optimized electrode materials containing Li2MnO3 along with two or three components [2,3,8e12,15e19]. In this work, Li2MnO3 has been integrated with another Mn-based layered cathode LiMnO2 in order to have lithium rich layered-layered solid solutions. LiMnO2 crystallizes in two
forms: orthorhombic (o-LiMnO2, Pmnm Symmetry) andmonoclinic (m-LiMnO2, C2/m Symmetry) and shows a high theoretical capacity of around 290 mAhg1 and better cyclability when operated be- tween 2.0 and 4.5 V vs. Liþ/Li [9,14,20].
The motive behind the present work is to study the physico- chemical and electrochemical performances of the lithium richMn- based layered-layered solid solutions of Li2MnO3-LiMnO2. The research work reported on solid solution in this investigation has not been explored in the past. Although, there are reports on various solid-solution system which have been investigated pre- viously like Li2MnO3-LiMO2 (M ¼ Ni1/3Mn1/3Co1/3, Ni1/2Mn1/2) [21], Li1þx(Ni0.3Mn0.7-yTiy)1xO2, 0 < x < 1/3, y ¼ 0, 0.1, 0.2 [22], xLi2MnO3(1x) LiNi2/3Co1/6Mn1/6O2 [23], 0.5Li2MnO3- 0.5Li(Mn0.375Ni0.375Co0.25O2) [24] etc. Apart from these reported works several attempts has also beenmade to synthesize and study electrochemical performances of some Li-rich manganese-based compositions using different methods [25,26] and morphology [27e29] etc. All the synthesized compositions in this work have been prepared using the sol-gel route and were examined using X- ray diffraction (XRD) for phase analysis, scanning electron micro- scopy (SEM) for microstructural investigation and Fourier- transform infrared spectroscopy (FTIR) for vibrational modes analysis. Electrochemical characterizations of as-prepared compo- sitions were carried out to study electrochemical performance and lithium ion kinetics.
2. Experimental
Stoichiometric amounts of lithium acetate dihydrate (CH3COO- Li.2H2O, Sigma-Aldrich, ~99%) and Manganous acetate ((CH3COO)2 Mn.4H2O, Fisher Scientific, 99%) were used as precursors for the development of solid solutions of Li2MnO3-LiMnO2 [Li(Li(1x)/
3Mn(xþ2)/3)O2; x ¼ 0.0, 0.1, 0.3 and 0.5] using sol-gel technique. Citric acid (Sigma-Aldrich,99.5%) was used as a chelating agent. A dispersed solution of all the precursors in deionized water was prepared with continuous stirring and then added to citric acid solution one by one. The pH value of the resulting solution was maintained between 6 and 7 using ammonia solution. The resulting solution of citric acid and precursors was heated at 70 C at 400 rpm with continuous stirring until a transparent gel was ob- tained. The gel thus obtained was dried at 120 C for 5 h in a vac- uum oven to obtain xerogel. The yielded powders were reground together using pestle mortar and again heated at 600 C for 3 h in air and then finally calcined at 900 C for 12 h in air and then allowed to cool normally up to room temperature. The powders thus obtained for x ¼ 0.0, 0.1, 0.3 and 0.5 were abbreviated as LMO, LMO1, LMO3 and LMO5, respectively. The heating rate of the furnace was kept at 5 C min1.
2.2. Material characterization
Synthesized samples were characterized using RIGAKUmake X- ray diffractometer coupled with CuKa1 radiation (l ¼ 1.540 Å) for quantitative analysis of crystal structure and phase composition of the synthesized structures. The XRD data was collected over an angular range of 10e70 with a step size of 0.02. Rietveld refine- ment of LMO sample was performed using ‘X'pert high score plus' software in order to obtain quantitative information about struc- tural parameters. The microstructural characterization of the pre- pared samples was carried out using HITACHI make S-3700N scanning electronmicroscope working at 15 kV. The as-synthesized samples weremixedwith 2.5 wt% of polyvinyl alcohol (PVA) binder, pressed into round pellets with 10 mm diameter by applying
pressure of 5 tons using a hydraulic press and then heated at 250 C for 2 h in air to remove PVA. In order to make electrical contacts, silver (Ag) paste was applied on both sides of pellets and then dried at 150 C for 1 h in air to remove moisture content. A two-point probe setup (AGILENT 4284A) was used to calculate the elec- tronic conductivity of the synthesized samples. Conductivity measurements were done at a bias voltage of 1.0 V and the fre- quency range was kept from 20 Hz to 1 MHz. The electronic con- ductivity (s) can be calculated using the following equation;
s ¼ 1 R
(2)
where R is the resistance of the sample (U), L and A is thickness (cm) and cross-sectional area (cm2) of the circular disc, respectively.
2.3. Electrode preparation and electrochemical characterization
For the electrochemical studies, two electrode coin half-cells (CR2032) were assembled. The working electrodes were prepared by dissolving 70wt% active material, 15 wt% super P as a conductive carbon and 15 wt% polyvinylidene difluoride (PVDF) as a binder in N-methyl-2-pyrrolidinone (NM2P) as a solvent. This results in the uniform mixing of the active materials which were stirred for 10 h. Thus obtained slurry was uniformly coated on Al foil current col- lector using the “doctor blade” technique and further dried out at 80 C in a vacuum oven for overnight to evaporate the solvent. Then circular electrodes with diameter 19 mm were punched and weighed. The electrodes were then dried up under vacuum at 60 C and finally brought inside the Mbraun make glove box having O2 and H2O level 0.5 ppm. The coin cells (CR2032) were assembled for electrochemical studies of the synthesized samples with metallic lithium as both counter and reference electrode and a microporous polypropylene sheet Celgard 2400 used as a separator. The electrolyte used was 1 M LiPF6 in EC: DMC 1:1 v/v. Cyclic vol- tammetry (CV) experiments were carried out over the potential range of 2.0e4.8 V (vs. Li/Liþ metal). The galvanostatic charge/ discharge (GCD) performance test, rate capability at different C- rates and Electrochemical Impedance Spectroscopy (EIS) using an AC voltage pulse of 5 mV in the frequency range of 10 mHz-10 kHz were performed. All the electrochemical tests were performed using EZstat-pro Nuvant systems Inc. (U.S.A) cell test system at 298 K.
3. Results and discussion
3.1. Structural and morphological analysis
The XRD patterns of the as-synthesized compositions are shown in Fig. 1. The pristine Li2MnO3 (LMO) shows diffraction peaks which correspond to the m-Li2MnO3 and could be assign to C2/m space group (Reference Code: 01-084-1634). The sharp and intense peaks indicate the crystalline nature of all the synthesized compositions. Even composite solid-solutions i.e. Li[Li(1x)/3Mn(xþ2)/3]O2 (x ¼ 0.1, 0.3 and 0.5) still shows the main characteristics peaks of m- Li2MnO3, indicating that the layered structure of the synthesized samples is still preserved. The (001) diffraction peaks of the syn- thesized compositions (shown in the magnified XRD image) clearly shows slight shifts towards higher 2q angle suggesting a minor shrink of the unit cell and decrease in lattice parameters. On the other hand, weak diffraction peaks located around 20e30 can be attributed to the superstructure ordering of the lattice which is caused by the arrangement of Mn ion into Li-Mn layers and are observed for Li-rich based materials [1,28,30e32]. However, the
Fig. 1. XRD patterns of solid solution compositions of Li2MnO3-LiMnO2.
Fig. 2. Rietveld refined XRD pattern of pristine Li2MnO3 (LMO).
R. Saroha et al. / Journal of Alloys and Compounds 696 (2017) 580e589582
intensity of these superlattice peaks become weaker on changing the value of x from 0.1 to 0.5 which indicates that the local ordering of Li/Mn has been reduced. In addition, the peak around 37 splits into two peaks i.e. (2 01) and (130), which can also be indexed to C2/ m space group with a high degree of crystallinity. Similarly, the peak around 45 also depicts the same results. However, the in- tensity of these doublets gets faded with an increase in x. The splitting of peaks located at 59 and 65 for all values of x has also been noticed which indicates the presence of spinel phase [15]. Rietveld refinement was performed for the pristine LMO synthe- sized sample to obtain quantitative structural information of the crystal lattice structure using ‘X'pert high score plus' software. The observed XRD pattern along with calculated and difference pattern for original LMO sample are shown in Fig. 2. No background sub- traction and data smoothening were done before refinement. The starting atomic position coordinates for pristine LMO were taken from somewhere else [33]. The structural refinement results indi- cate that the goodness of fitting factor i.e. c2 for primitive LMO is around 4.5% which demonstrates that the results are reliable. The crystal structure generated after refinement has been displayed in the inset of Fig. 2 where Li and Mn atoms are surrounded by O atoms to form LiO6 and MnO6 octahedra. The lattice parameters obtained from refinement for bare LMO are: a ¼ 4.928 Å, b ¼ 8.522 Å and c ¼ 5.022 Å. The average crystallite size for pristine LMO was calculated to be 54 nm using Scherrer's formula d ¼ kl/b cos q, where d is the mean crystallite size, k is the constant parameter of shape factor (0.94), l is the wavelength of X-ray (0.154 nm), b is the full-width half maximum corresponding to (001) peak and q is the corresponding angle of Bragg diffraction. These results suggest that the variation in x not only increase the crystallinity of the synthesized samples but also induce phase transformation to some extent.
The effect of variation in x value on the morphologies of Li [Li(1x)/3Mn(xþ2)/3]O2 (x ¼ 0.0, 0.1, 0.3 and 0.5) composites are shown in Fig. 3(a)-(d). All prepared samples display nearly same morphology with plate-like and relatively spherical grains. The
average grain size of LMO, LMO1, LMO3 and LMO5 composites were found to be around 0.45, 0.3, 0.5 and 0.45 mm, respectively. Therefore, the average grain sizes of the composites were distrib- uted in a range of 300e450 nm. SEM images also show that the synthesized nanocomposites possess dispersed grains with little aggregation.
Among all synthesized samples, the pristine LMO and LMO3 compositions are chosen to investigate the morphological proper- ties further. Fig. 4(A) and (B) shows the TEM/HRTEM images of the LMO and LMO3 synthesized samples. TEM image of pristine LMO (Fig. 4(A)) sample indicates that the particles are homogeneously
Fig. 3. SEM micrographs of Li2MnO3-LiMnO2 synthesized compositions (a) LMO; (b) LMO1; (c) LMO3; (d) LMO5.
R. Saroha et al. / Journal of Alloys and Compounds 696 (2017) 580e589 583
distributed with little agglomeration. It can also be observed from TEM image that the particles are nearly spherical and possess nearly uniform size. The average particle size was calculated using ImageJ software and observed to be 170 nm for LMO composition. HRTEM image (inset of Fig. 4(A)) clearly indicates that the sample is well crystallinewith uniformly spaced lattice fringes. The interlayer spacing of pristine LMO sample was observed to be 0.472 nmwhich corresponds to (001) plane of the monoclinic structure of synthe- sized LMO composition. Likewise, TEM image oF LMO3 composi- tion (Fig. 4(B)) also reveals the spherical morphology of the synthesized composition with slight agglomeration. The average particle size for LMO3 composition was calculated to be 220 nm. HRTEM image (inset of Fig. 4(B)) for LMO3 composition also reveals the crystalline nature of the sample with uniformly spaced lattice fringes. The interlayer spacings of LMO3 compositionwere found to be 0.470 and 0.475 nm which again corresponds to (001) plane of the monoclinic structure. The obtained TEM/HRTEM results are well synchronized with XRD and SEM results.
3.2. Cation environment study-FTIR
FTIR spectroscopy proves to be an appropriate and dominant technique for the structural evaluation and to study local envi- ronment of the cations in the crystal lattice of close-packed oxygen atoms. Vibrational modes are generally assigned to the motion of cations with respect to an oxygen atom. The local cation environ- ment of synthesized compositions in the range of 400e600 cm1
has been studied using FTIR spectroscopy and results are displayed in Fig. 5. It can be observed from FTIR graph that the spectrum of all synthesized samples shows nearly similar features. In the spectrum of LMO, four bands located around 520 cm1 (weak band), 465 cm1 (slightly broad peak), 430 cm1 (sharp band) and at 417 cm1 (less intense band) can be observed. These four IR active modes are assigned to the stretching and bending vibrations of octahedrally coordinated molecule of MnO6 in Li[Li(1x)/3Mn(xþ2)/3] O2 matrix [32,34]. The weak band located at 520 cm1 slightly
moves towards higher wavenumber side for LMO1, LMO3 and LMO5 compositions which may be attributed to the variation in Li and Mn amount in Li[Li(1x)/3Mn(xþ2)/3]O2 solid solution which in- duces a difference in the local cation environment. Similarly, the band at 412 cm1 shift towards lower wavenumber side for LMO3 and LMO5which confirms that the local environment of the cations has been changed.
3.3. Electronic conductivity measurements
Electronic conductivity gives the direct measurement of the electron kinetics and hence, the electrochemical properties of the material. Fig. 6 displays the AC electrical resistance measurement for the as-synthesized compositions. The frequency range for the data collection was kept from 20 Hz to 1 MHz. From Fig. 6 it can be seen that all the compositions display depressed semicircle whose intercept at x-axis gives the AC resistance of the cathode active material. It can be seen that excluding LMO3 sample all the com- positions have AC resistance in megaohm. The AC conductivity values for the samples LMO, LMO1, LMO3 and LMO5 were found to be 10.1 109, 10.8 109, 1.65 107 and 9.36 109 S/cm, respectively. Therefore, among all the synthesized samples LMO3 shows the highest AC conductivity and hence, superior electro- chemical performances are expected from LMO3 composition as compared to other synthesized samples.
3.4. Electrochemical performance of Li2MnO3-LiMnO2 solid solutions
Cyclic voltammetry measurements give an insight into the electrochemical reactions taking place during the charging- discharging process. Fig. 7 displays the CV results of the as- synthesized samples in the voltage window of 2.0e4.8 V. The data for the CV measurement was collected at a scan rate of 0.05 mVs1. It can be observed from Fig. 6 that pristine LMO possess weak anodic peak around 4.7 V and no symmetric cathodic peak. It
Fig. 4. TEM and HRTEM images of the synthesized compositions (A) LMO; (B) LMO3.
Fig. 5. FTIR spectrum of the synthesized compositions in the wavenumber range of 400e600 cm1.
R. Saroha et al. / Journal of Alloys and Compounds 696 (2017) 580e589584
indicates the loss of lattice oxygen along with lithium extraction frommonoclinic crystal structure as given by eq. (1) [33]. Therefore, CV result indicates that the activation of primitive LMO will take place at 4.7 V and this process will not be reversible. However, for LMO1, LMO3 and LMO5 compositions the intensity of the peak at 4.7 V decreases significantly which indicates that the redox reac- tionmechanism is different from LMO.Moreover, MnO2 obtained in eq. (1) is assumed to be electrochemically active and formed spinel structure during Li-intercalation. The samples with x > 0 i.e. LMO1, LMO3 and LMO5 shows a redox couple around 3.0/2.8 V, 4.1/4.0 V and anodic peak around 2.1 V. The redox couple at 4.1/4.0 can be attributed to insertion of lithium ion at vacant 8a tetrahedral sites of the spinel structure formed by MnO2 [9]. Likewise, the emer- gence of redox couple at 3.0/2.8 V is due to intercalation of Li-ion at 16c octahedral sites of the spinel structure [33]. The origin of anodic peak around 2.1 V can be attributed to structural modification of the material which generally take place due to high upper voltage limit and can be assigned to spinel Mn4þ/Mn3þ redox couple [35]. When charged at higher voltage, transition metal dissociation takes place which leads to phase transformation from layered to spinel structure. However, the intensity of these redox couples was found
to vary with x. This phenomenon may be correlated to the man- ganese dissolution which originated due to Jahn-Teller distortion and leads to change in oxidation state of Mn in the composition [9]. The average oxidation states of Mn in LMO, LMO1, LMO3 and LMO5 were calculated to be 4.0, 3.85, 3.61 and 3.40, respectively accord- ing to the redox titration method [27]. As average oxidation state of Mn changes from 4.0 to 3.4, Jahn-Teller distortion leads to disso- ciation of Mn3þ to Mn4þ andMn2þ which results in active mass loss and a decrease in intensity [9]. This change of oxidation state is generally expressed in terms of the transformation of the layer-to- spinel structure as predicted by XRD (splitting of peaks at 2q ¼ 59
and 65). The presence of shoulder peaks at 4.04, 3.98, 3.93 and 3.96 V also indicates the existence of small spinel phase due to electrochemical active MnO2. Therefore, the above results clearly indicate the coexistence of spinel structure along with layered structure while the redox process observed at 2.8 V and 4.1 V for LMO1, LMO3 and LMO5 samples will lead to stable capacity and cyclic performances.
Fig. 8 shows the initial charge-discharge capacity curves of Li [Li(1x)/3Mn(xþ2)/3]O2 (x ¼ 0.0, 0.1, 0.3 and 0.5) solid solution at a current density of 10 mA/g and in a voltage window of 4.8e2.0 V. The initial discharge capacities for composites LMO, LMO1, LMO3 and LMO5 were found to be 81 ± 5, 97 ± 5, 177 ± 5 and 71 ± 5 mAhg1, respectively. It can be observed that as x value in- creases capacity increases, goes maximum for LMO3 and then de- creases for LMO5. For all composite samples, a sloping voltage plateau above 4.5 V can be seen in the charging curves. However, no discharge plateau was observed above 4.5 V for the composite samples which is associated with the oxidation of O2- ions and electrochemical activation of Li2MnO3 by irreversible removal of oxygen from the lattice. Moreover, the charge-discharge profile of the LMO1, LMO3 and LMO5 samples exhibits three voltage plateaus around 4.25 V, 2.87 V and 2.25 V. The voltage plateaus around 4.25 V and 2.87 V originate due to change in oxidation state of Mn ion during charging-discharging. The length of voltage plateau at 4.25 V, 2.87 V and 2.25 V increases with increase in the value of x, reaching amaximum for LMO3 and then decreases for LMO5which is possibly due to Jahn-Teller distortion as predicted in CV results. Among all the four samples, the LMO3 composition gives the highest discharge capacity and the composition LMO5 delivers the
Fig. 6. AC conductivity measurements of synthesized solid solutions in the frequency range of 20 Hze1MHz.
Fig. 7. Cyclic Voltammograms of the Li(Li(1x)/3Mn(xþ2)/3)O2 samples at a scan rate of 0.05 mVs1. Fig. 8. 1st charge/discharge curves of solid solution samples at a current density of
10 mA/g (icharge ¼ idischarge).
R. Saroha et al. / Journal of Alloys and Compounds 696 (2017) 580e589 585
least charge-discharge capacity. These results indicate superior lithium ion kinetics and excellent reactivity for LMO3 sample. Fig. 9(a)-(d) shows the charge/discharge capacity curves of LMO, LMO1, LMO3 and LMO5, respectively for 1st, 2nd, 10th and 50th cycle at 10 mA/g. From Fig. 9(a), it can be observed that discharge capacity of LMO composition at the 50th cycle is much greater as compared to 1st cycle. This is due to activation of LMO compound due to the removal of O2 along with lithium extraction which causes an increase in the length of voltage plateau around 3.2 V.
However, for LMO1 (Fig. 9(b)), LMO3 (Fig. 9(c)) and LMO5 (Fig. 9(d)) compositions the charge/discharge profiles for 1st, 2nd, 10th and 50th cycle remains nearly same. These results indicate that the integrity of the synthesized compositions remains unchanged during cycling.
Fig. 10(a)-(d) shows the charge/discharge curves of the synthe- sized samples at different current densities (10e1000mA/g) for the initial cycle. It can be observed from Fig. 10 that among all the synthesized samples LMO3 (Fig. 10(c)) delivers the highest specific
Fig. 9. Charge/Discharge profiles of the prepared compositions for 1st, 2nd, 10th and 50th cycle at 10 mA/g: (a) LMO; (b) LMO1; (c) LMO3; (d) LMO5.
Fig. 10. Initial charge/discharge curves of the synthesized compositions at various current densities: (a) LMO; (b) LMO1; (c) LMO3; (d) LMO5.
R. Saroha et al. / Journal of Alloys and Compounds 696 (2017) 580e589586
Fig. 11. Performance of Li(Li(1x)/3Mn(xþ2)/3)O2 solid solution samples: (A) Rate per- formance at different current densities (icharge ¼ idischarge); (b) Cycling performance at current density of 10 mA/g for 120 charge/discharge cycles.
Fig. 12. Rate capability of the synthesized compositions as a function of x in Li(Li(1x)/
3Mn(xþ2)/3)O2.
R. Saroha et al. / Journal of Alloys and Compounds 696 (2017) 580e589 587
discharge capacity at all current densities. LMO3 shows an initial discharge capacity of 177 (±5) mAh/g at 10 mA/g and delivers nearly 37% of the capacity at a high current density of 1 A/g. On the other hand, LMO5 (Fig. 10(d)) shows the least specific discharge capacity at all current densities because of its low conductivity.
Fig. 11(a) shows the stepwise rate performance of the Li[Li(1x)/
3Mn(xþ2)/3]O2 compositions at current densities of 10e1000 mA/g (10 mA/g ¼ 0.05C). It can be observed that the composition LMO3 delivers the highest discharge capacity at all current densities. The initial discharge capacities for composites LMO, LMO1, LMO3 and
Table 1 Electrochemical performance of synthesized compositions at different C-rates and cyclin
Samples Ra (%)
0.05C (±5) (mAhg1) 0.1C 0.25C 0.5
LMO 81 144.4 128.3 11 LMO1 97 84.5 81.4 72 LMO3 177 84.7 70.6 62 LMO5 71 98.8 90.1 78
a Capacity retention ratios of samples at different C-rates as compared to 0.05C. b Capacity retention rates of samples compared to the 1st cycle at 0.05C.
LMO5 were found to be 81 ± 5, 97 ± 5, 177 ± 5 and 71 ± 5 mAhg1, respectively at 10 mA/g. However, the discharging capacity values for LMO were found to increase continuously from 81 mAhg1 for the 1st cycle to 133 mAhg1 for 5th discharge cycle which is due to the activation of Li2MnO3 as given by eq. (1). Moreover, LMO5 composite possesses the lowest discharge capacities even at low current densities but delivers fairly better discharge capacity values at high current rates as compared to the pristine LMO and LMO1 composites. Table 1 compares the discharge capacity values at each current density with respect to 10 mA/g rate. LMO3 displays highest discharge capacity and retains ~37% of its capacity value even at the high current rate of 5C (1A/g). The loss of capacity at high current densities can be linked to the large polarization effects of the composites which hindered lithium ion kinetics. Fig. 11(b) displays the cycling stability of the synthesized composites at 10 mA/g current density for 120 charge-discharge cycles at room temperature. It is evident from Fig.11(b)) and Table 1 that except for LMO3 composite sample, discharge capacity at 120th cycle comes out to be more as compared to 1st cycle for LMO, LMO1 and LMO5 composition. LMO3 shows an initial discharge capacity of 177(±5) mAhg1 at 10 mA/g and retains nearly 97% of its capacity after 120 charge/discharge cycles at same current density.
Fig. 12 depicts the dependence of the rate performance on the variation of x. From the Figure, it can be observed that at low C- rates, the difference in electrochemical properties is mainly due to the change in oxidation state of Mn ionwhich subsequently leads to manganese dissolution and inferior capacity performance. Under high C-rates, the alteration in discharge capacities is predominantly due to increase in polarization potential of the cathode active ma- terial which leads to poor discharge capacity values.
g capability at 0.05C-rate (10 mA/g).
5C 0.05C (±5) (mAhg1)
C 2.5C 1st 120th Rb (%)
6.0 55.5 6.1 81 144 177.7 .1 32.9 20.6 97 105 108.2 .1 50.8 37.2 177 172 97.1 .8 83.0 67.6 71 72 101.4
Fig. 13. (a) Impedance spectra of synthesized compositions measured at AC pulse of amplitude 5 mV; (b) Relationship plot between Z’ and u0.5 in low-frequency region.
Table 2 Electrode kinetic factors for synthesized solid solution compositions.
Sample Re (U) Rct (U) sw (U s0.5) D (cm2s1)
LMO 16.30 177.09 210.10 5.76 1017
LMO1 10.46 71.08 92.53 2.97 1016
LMO3 13.58 50.54 57.26 7.75 1016
LMO5 9.06 1059.6 837.09 3.63 1018
R. Saroha et al. / Journal of Alloys and Compounds 696 (2017) 580e589588
EIS of the fresh cells was also performed to study the variation in electrode reaction impedance of the prepared composition. Fig. 13(a) demonstrated the EIS curves of Li[Li(1x)/3Mn(xþ2)/3]O2 compositions in the frequency range of 10 mHz-10 kHz using an AC voltage pulse of 5 mV amplitude. It can be observed from Fig. 13(a) that all the EIS curves of the composites samples enclose a depressed semicircle lies in the high-frequency region and a straight line falls in the low-frequency region. The intercept by the semi-circle at the real axis in the high-frequency region gives the ohmic resistance (Rs) of the cell which usually represents the resistance of the electrode material and the elec- trolyte. The part of semicircle lies in the middle-frequency range is mainly attributed to the electrochemical reactions taking place at the electrode and electrolyte interface and named as charge transfer resistance (Rct). The 45 inclined line with the x-axis in the low-frequency region signifies the Warburg impedance (Zw) which is related to Li-ion diffusion in the electrolyte solution. The values of lithium ion diffusion coefficient in the bulk electrode materials can be calculated using the following equation [36]:
D ¼ 0:5R2T2
(3)
where D is Li-ion diffusion coefficient, R is gas constant, T is abso- lute temperature, A is electrode area, C is the lithium-ion concen- tration, F is the faraday constant and sW is the Warburg factor which is given by the following equation:
Z’¼ RsþRctþsWu0:5 (4)
Z’ is real part of the impedance and the Warburg factor can be obtained from the slope of Z’ and u0.5 curve in the low-frequency region. Fig. 13(b) demonstrates the plot between Z’ and u0.5 in the low-frequency region for all the synthesized composites samples and their values are listed in Table 2. From the EIS plot and Table 2, it is clear that LMO3 has minimum charge transfer resistance and therefore, highest lithium ion diffusion coefficient (7.75 1016
cm2s1) among all the compositions confirming the CV and charge/ discharge results. However, as compared to other manganese- based cathode materials, the lithium ion diffusion coefficients are found to be low [37,38]. This may be due to the presence of LiMnO2 component. LiMnO2 undergo irreversible structural changes from layered to spinel as given by equation (5) [14]:
3LiMnO2 þ 1 2 O2/LiMn2O4þLi2MnO3 (5)
Spinel material possesses low discharge capacities and small lithium ion diffusion coefficient as compared to the o-LiMnO2 [16,39]. Therefore, the presence of small traces of spinel phase induced by o-LiMnO2 might be the possible reason for low lithium ion diffusion coefficient and hence, relatively poor electrochemical performances. Also, the lithium-ion diffusion coefficient was found to be minimum for LMO5 composition and therefore, it possesses inferior electrochemical performances. The EIS results clearly indicate that the LMO3 composition possesses the lowest charge transfer resistance due to which it delivers the highest discharge capacities among all the synthesized samples.
4. Conclusions
A series of solid solution of layered-layered composites Li[Li(1x)/
3Mn(xþ2)/3]O2 (x ¼ 0.0, 0.1, 0.3 and 0.5) were successfully synthe- sized using sol-gel technique. XRD results confirm the coexistence of layered structure along with spinel structure. Microstructural investigations reveal that the average grain size of all the compo- sitions lies in a range of 300e450 nm. Among all the synthesized compositions LMO3 shows the best rate and cycling performances. It delivers an initial specific discharge capacity of 177± 5mAhg1 at a current density 10 mA/g. The better charge/discharge profiles and highest rate and cycling performances of the LMO3 composition can be attributed to highest electronic conductivity (1.65 107 S/ cm) and maximum lithium-ion diffusion coefficient (7.75 1016
cm2s1) factor.
R. Saroha et al. / Journal of Alloys and Compounds 696 (2017) 580e589 589
Acknowledgments
Authors are thankful to SERB and FIST, Department of Science and Technology (DST), New Delhi for financial support to carry out this work under the project grants Nos.: SR/FTP/PS-039/2011 and SR/FST/PSI-164/2011(C), respectively.
References
[1] Y. Sun, Y. Shiosaki, Y. Xia, H. Noguchi, The preparation and electrochemical performance of solid solutions LiCoO2eLi2MnO3 as cathode materials for lithium ion batteries, J. Power Sources 159 (2006) 1353e1359.
[2] D. Kim, J. Gim, J. Lim, S. Park, J. Kim, Synthesis of xLi2MnO3$(1x) LiMO2 (M¼ Cr, Mn, Co, Ni) nanocomposites and their electrochemical properties, Mater. Res. Bull. 45 (2010) 252e255.
[3] C. Park, S. Kim, I.R. Mangani, J. Lee, S. Boo, J. Kim, Synthesis and materials characterization of Li2MnO3eLiCrO2 system nanocomposite electrode mate- rials, Mater. Res. Bull. 42 (2007) 1374e1383.
[4] M.M. Thackeray, Y. Shao-Horn, A.J. Kahaian, K.D. Kepler, E. Skinner, J.T. Vaughey, S.A. Hackney, Structural fatigue in spinel electrodes in high voltage (4 V) Li/LixMn2O4 cells, Electrochem. Solid-State Lett. 1 (1998) 7e9.
[5] J. Li, Q. Qu, L. Zhang, L. Zhang, H. Zheng, A monodispersed nano-hexahedral LiFePO4 with improved power capability by carbon-coatings, J. Alloys Compd. 579 (2013) 377e383.
[6] Q. Liu, W. Liu, D. Li, H. Chen, Z. Wang, LiFe1x(Ni0.98Co0.01Mn0.01)xPO4/C (x¼ 0.01, 0.03, 0.05, 0.07) as cathode materials for lithium-ion batteries, Electro- chim. Acta 184 (2015) 143e150.
[7] S.-X. Zhao, H. Ding, Y.-C. Wang, B.-H. Li, C.-W. Nan, Improving rate perfor- mance of LiFePO4 cathode materials by hybrid coating of nano-Li3PO4 and carbon, J. Alloys Compd. 566 (2013) 206e211.
[8] Y. Wu, A. Manthiram, Structural stability of chemically delithiated layered (1z)Li[Li1/3Mn2/3]O2ezLi[Mn0.5yNi0.5yCo2y]O2 solid solution cathodes, J. Power Sources 183 (2008) 749e754.
[9] Q. Zhang, T. Peng, D. Zhan, X. Hu, Synthesis and electrochemical property of xLi2MnO3$(1x)LiMnO2 composite cathode materials derived from partially reduced Li2MnO3, J. Power Sources 250 (2014) 40e49.
[10] D. Mori, H. Sakaebe, M. Shikano, H. Kojitani, K. Tatsumi, Y. Inaguma, Synthesis, phase relation and electrical and electrochemical properties of ruthenium- substituted Li2MnO3 as a novel cathode material, J. Power Sources 196 (2011) 6934e6938.
[11] S. Kim, C. Kim, J.-K. Noh, S. Yu, S.-J. Kim, W. Chang, W.C. Choi, K.Y. Chung, B.- W. Cho, Synthesis of layeredelayered xLi2MnO3$(1x)LiMO2 (M¼ Mn, Ni, Co) nanocomposite electrodes materials by mechanochemical process, J. Power Sources 220 (2012) 422e429.
[12] W. West, R. Staniewicz, C. Ma, J. Robak, J. Soler, M. Smart, B. Ratnakumar, Implications of the first cycle irreversible capacity on cell balancing for Li2MnO3eLiMO2 (M¼ Ni, Mn, Co) Li-ion cathodes, J. Power Sources 196 (2011) 9696e9701.
[13] J. Li, S. Jeong, R. Kloepsch, M. Winter, S. Passerini, Improved electrochemical performance of LiMO2 (M¼ Mn, Ni, Co)eLi2MnO3 cathode materials in ionic liquid-based electrolyte, J. Power Sources 239 (2013) 490e495.
[14] J. Molenda, M. Ziemnicki, J. Marzec, W. Zajac, M. Molenda, M. Bucko, Elec- trochemical and high temperature physicochemical properties of ortho- rhombic LiMnO2, J. Power Sources 173 (2007) 707e711.
[15] C. Yu, G. Li, X. Guan, J. Zheng, L. Li, T. Chen, Composites Li2MnO3$LiMn1/3Ni1/ 3Co1/3O2: optimized synthesis and applications as advanced high-voltage cathode for batteries working at elevated temperatures, Electrochim. Acta 81 (2012) 283e291.
[16] Z. Su, Z. Lu, X. Gao, P. Shen, X. Liu, J. Wang, Preparation and electrochemical properties of indium-and sulfur-doped LiMnO2 with orthorhombic structure as cathode materials, J. Power Sources 189 (2009) 411e415.
[17] K.G. Gallagher, S.-H. Kang, S.U. Park, S.Y. Han, xLi2MnO3$(1x) LiMO2 blended with LiFePO4 to achieve high energy density and pulse power capability, J. Power Sources 196 (2011) 9702e9707.
[18] J.-H. Lim, H. Bang, K.-S. Lee, K. Amine, Y.-K. Sun, Electrochemical character- ization of Li2MnO3eLi[Ni1/3Co1/3Mn1/3]O2eLiNiO2 cathode synthesized via co- precipitation for lithium secondary batteries, J. Power Sources 189 (2009) 571e575.
[19] H. Shu, X. Wang, Q. Wu, B. Hu, X. Yang, Q. Wei, Q. Liang, Y. Bai, M. Zhou, C. Wu, Improved electrochemical performance of LiFePO4/C cathode via Ni and Mn co-doping for lithium-ion batteries, J. Power Sources 237 (2013) 149e155.
[20] E.M. Jin, B. Jin, Y.-S. Jeon, K.-H. Park, H.-B. Gu, Electrochemical properties of LiMnO2 for lithium polymer battery, J. Power Sources 189 (2009) 620e623.
[21] L. Song, Z. Tang, Y. Chen, Z. Xiao, L. Li, H. Zheng, B. Li, Z. Liu, Structural analysis of layered Li2MnO3eLiMO2 (M¼ Ni1/3Mn1/3Co1/3, Ni1/2Mn1/2) cathode mate- rials by Rietveld refinement and first-principles calculations, Ceram. Int. 42 (2016) 8537e8544.
[22] M. Tabuchi, H. Kageyama, K. Takamori, Y. Imanari, K. Nakane, Synthesis and electrochemical characterization of Ni-and Ti-substituted Li2MnO3 positive electrode material using coprecipitationehydrothermalecalcination method, Electrochim. Acta 210 (2016) 105e110.
[23] J. Shojan, V.R. Chitturi, J. Soler, O. Resto, W.C. West, R.S. Katiyar, High energy xLi2MnO3e(1x) LiNi2/3Co1/6Mn1/6O2 composite cathode for advanced Li-ion batteries, J. Power Sources 274 (2015) 440e450.
[24] J. Patra, P.P. Dahiya, C.-J. Tseng, J. Fang, Y.-W. Lin, S. Basu, S.B. Majumder, J.- K. Chang, Electrochemical performance of 0.5Li2M- nO3e0.5Li(Mn0.375Ni0.375Co0.25)O2 composite cathode in pyrrolidinium-based ionic liquid electrolytes, J. Power Sources 294 (2015) 22e30.
[25] C. Hou, Y. Oaki, E. Hosono, H. Lin, H. Imai, Y. Fan, F. Dang, Bio-inspired syn- thesis of xLi2MnO3-(1x)LiNi0.33Co0.33Mn0.33O2 lithium-rich layered cathode materials, Mater. Des. 109 (2016) 718e725.
[26] A. Vu, Y. Qin, C.-K. Lin, A. Abouimrane, A.K. Burrell, S. Bloom, D. Bass, J. Bare~no, I. Bloom, Effect of composition on the voltage fade phenomenon in lithium-, manganese-rich xLiMnO3$(1x)LiNiaMnbCocO2: a combinatorial synthesis approach, J. Power Sources 294 (2015) 711e718.
[27] F. Yang, Q. Zhang, X. Hu, T. Peng, Synthesis of layered xLi2MnO3$(1x)LiMnO2 nanoplates and its electrochemical performance as Li-rich cathode materials for Li-ion battery, Electrochim. Acta 165 (2015) 182e190.
[28] B. Li, Y. Yu, J. Zhao, Facile synthesis of spherical xLi2MnO3$(1x)Li(Mn0.33- Co0.33Ni0.33)O2 as cathode materials for lithium-ion batteries with improved electrochemical performance, J. Power Sources 275 (2015) 64e72.
[29] G. Xu, J. Li, X. Li, H. Zhou, X. Ding, X. Wang, F. Kang, Understanding the electrochemical superiority of 0.6Li[Li1/3Mn2/3]O2-0.4Li[Ni1/3Co1/3Mn1/3]O2 nanofibers as cathode material for lithium ion batteries, Electrochim. Acta 173 (2015) 672e679.
[30] M. Tabuchi, Y. Nabeshima, T. Takeuchi, K. Tatsumi, J. Imaizumi, Y. Nitta, Fe content effects on electrochemical properties of Fe-substituted Li2MnO3 positive electrode material, J. Power Sources 195 (2010) 834e844.
[31] C. Miao, P. Bai, Q. Jiang, S. Sun, X. Wang, A novel synthesis and character- ization of LiFePO4 and LiFePO4/C as a cathode material for lithium-ion battery, J. Power Sources 246 (2014) 232e238.
[32] X. Dong, Y. Xu, L. Xiong, X. Sun, Z. Zhang, Sodium substitution for partial lithium to significantly enhance the cycling stability of Li2MnO3 cathode material, J. Power Sources 243 (2013) 78e87.
[33] S. Basu, P.P. Dahiya, M. Akhtar, S.K. Ray, J.K. Chang, S.B. Majumder, High en- ergy density layered-spinel hybrid cathodes for lithium ion rechargeable batteries, Mater. Sci. Eng. B 213 (2016) 148e156.
[34] Y. Li, Y. Makita, Z. Lin, S. Lin, N. Nagaoka, X. Yang, Synthesis and character- ization of lithium manganese oxides with core-shell Li4Mn5O12@Li2MnO3 structure as lithium battery electrode materials, Solid State Ionics 196 (2011) 34e40.
[35] J. Pires, L. Timperman, A. Castets, J.S. Pe~na, E. Dumont, S. Levasseur, R. Dedryvere, C. Tessier, M. Anouti, Role of propane sultone as an additive to improve the performance of a lithium-rich cathode material at a high po- tential, RSC Adv. 5 (2015) 42088e42094.
[36] J. Lin, D. Mu, Y. Jin, B. Wu, Y. Ma, F. Wu, Li-rich layered composite Li [Li0.2Ni0.2Mn0.6]O2 synthesized by a novel approach as cathode material for lithium ion battery, J. Power Sources 230 (2013) 76e80.
[37] P. Suresh, A.K. Shukla, N. Munichandraiah, Characterization of Zn-and Fe- substituted LiMnO2 as cathode materials in Li-ion cells, J. Power Sources 161 (2006) 1307e1313.
[38] K.M. Shaju, G.V.S. Rao, B.V.R. Chowdari, EIS and GITT studies on oxide cath- odes, O2-Li(2/3)þx(Co0.15Mn0.85)O2 (x¼ 0 and 1/3), Electrochim. Acta 48 (2003) 2691e2703.
[39] H. Zhao, S. Liu, X. Liu, M. Tan, Z. Wang, Y. Cai, S. Komarneni, Orthorhombic LiMnO2 nanorods as cathode materials for lithium-ion batteries: synthesis and electrochemical properties, Ceram. Int. 42 (2016) 9319e9322.
1. Introduction
2. Experimental
3. Results and discussion
3.2. Cation environment study-FTIR
3.3. Electronic conductivity measurements
4. Conclusions