Structural stability and superior electrochemical performance of Sc-doped LiMn2O4 spinel as cathode for lithium ion batteries

Accepted Manuscript Structural stability and superior electrochemical performance of Sc-doped LiMn2O4 spinel as cathode for lithium ion batteries Subramani Bhuvaneswari, U.V. Varadaraju, R. Gopalan, Raju Prakash PII:

S0013-4686(19)30173-2

DOI:

https://doi.org/10.1016/j.electacta.2019.01.174

Reference:

EA 33577

To appear in:

Electrochimica Acta

Received Date: 29 November 2018 Revised Date:

17 January 2019

Accepted Date: 26 January 2019

Please cite this article as: S. Bhuvaneswari, U.V. Varadaraju, R. Gopalan, R. Prakash, Structural stability and superior electrochemical performance of Sc-doped LiMn2O4 spinel as cathode for lithium ion batteries, Electrochimica Acta (2019), doi: https://doi.org/10.1016/j.electacta.2019.01.174. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

GRAPHICAL ABSTRACT Sc-doped LiMn2O4 spinel by hydrothermal and solid state method shows significant enhancement in cyclic stability and rate capability performance than pristine spinel as

AC C

EP

TE D

M AN U

SC

RI PT

cathode in lithium ion batteries.

ACCEPTED MANUSCRIPT

Structural stability and superior electrochemical performance of Sc-doped LiMn2O4 spinel as cathode for lithium ion batteries#

a

RI PT

Subramani Bhuvaneswaria,b U. V. Varadarajub R. Gopalana Raju Prakasha* Centre for Automotive Energy Materials, International Advanced Research Centre for powder

Metallurgy and new Materials (ARCI), Taramani, Chennai - 600 113, India

Department of Chemistry, Indian Institute of Technology Madras, Chennai-600 036, India

SC

b

*Corresponding author: [email protected] Doped spinel compound Part-1

M AN U

#

ABSTRACT

Scandium doped LiScxMn2-xO4 compounds are synthesized by solid-state method, which show single phase with rod-like polyhedron morphology. The Sc-doping decreases the lattice parameter ‘a’

TE D

marginally due to the change in the inter-atomic distance of the metal oxide bonds as confirm by Rietveld refinement. In addition, the expansion of LiO4 tetrahedron and contraction of MnO6 octahedron by ~0.01Å upon doping are observed. The Sc2p3/2 peak at 402.5 eV and Sc2p1/2 peak at

EP

407.2 eV in the XPS spectrum of LiSc0.06Mn1.94O4 confirms the presence of Sc in the spinel structure. The symmetric stretching of Mn−O bond of LiSc0.06Mn1.94O4 shifts lower value (~4 cm−1) than that of

AC C

LiMn2O4 indicating the occupancy of Sc3+ ion in the octahedral site. The diffusion coefficient value of LiSc0.06Mn1.94O4 (1×10−12cm2s−1) is one-order higher than that of undoped LiMn2O4 (1×10−13 cm2s−1). LiMn2O4 delivers a discharge capacity of 117 mAhg−1 at 1C with a capacity retention of 74% after 500 cycles, whereas under similar condition LiSc0.06Mn1.94O4 delivers a discharge capacity of 114 mAhg−1 with a capacity retention of >90%. LiSc0.06Mn1.94O4 also delivers excellent rate capability due to high diffusion coefficient and less charge transfer resistance compared to the parent compound. The structure and morphology of the Sc-doped electrode after 500 cycles remains intact without any 1

ACCEPTED MANUSCRIPT

formation of Mn-rich agglomeration suggest that the reduction of Mn2+ ion dissolution as well as JahnTeller distortion. Hence LiSc0.06Mn1.94O4 can be a potential cathode material for lithium ion batteries.

RI PT

Keywords: LiMn2O4, Scandium doping, lithium ion batteries, cyclic stability, rate capability

1 Introduction

SC

High energy and power density cathode materials are required for Lithium ion Batteries (LIBs) for Electric Vehicles (EVs) and Hybrid Electric Vehicles (HEVs) applications [1−6]. LiMn2O4 (LMO)

M AN U

spinel has been widely studied as a cathode for LIBs because of its high voltage, high power, low cost and environmental benign [7−9]. LiMn2O4 has a working voltage of 4.1 V (vs. Li/Li+) with a theoretical capacity of 148 mAhg−1 [10, 11]. It has a cubic crystal structure with Fd3m space group (227), where, Li atoms resides in the 8a tetrahedral sites, Mn atoms resides in the 16d octahedral sites and oxygen atom resides in the 32e sites in the spinel crystal lattice. Remaining empty voids are

TE D

interconnected and give a 3D diffusion pathway for fast lithium ion diffusion kinetics. Thus, the 3D diffusion increases the high power density of LMO spinel. Despite that LMO exhibits significant

EP

capacity fade in the cycling process especially at elevated temperature (≥ 55°C) which is mainly due to (i) the formation of tetragonal Li2Mn2O4 phase on the surface of LiMn2O4 and the associated Jahn–

AC C

Teller distortion, the presence of secondary phase Li2Mn2O4 in the electrode hinders the Li-ion diffusion kinetics results in huge volume change and severe capacity fade at the deeply discharged state [12, 13]. (ii) slow dissolution of Mn2+ from LMO electrode into the electrolyte due to the disproportionation reaction of 2Mn3+ → Mn4+ +Mn2+ [14,15] (iii) formation of new phases during charge state deteriorates the spinel phase [16]. Several strategies have been followed to suppress the capacity fade in the spinel; very recently a few research groups have reported improved electrochemical performances using octahedral and truncated spinel LMO nanoparticles [17, 18]. 2

ACCEPTED MANUSCRIPT

Coating of metal oxides on the surface of LMO also showed the enhanced performance by minimizing side reactions on the electrode surface during charging/discharging [19, 20]. Doping of transition and

RI PT

rare earth metal cations (Li, Cr, Co, Ni, Fe, Al, Sc, Nd, Gd, Dy, In and Y) [21−31], and anions (F, S), [32, 33] in LiMn2-xMxO4 are the most commonly used methods to improve the structural stability and charge transport property of the spinel. The doping of transition metal ions and electrochemically inactive rare earth metal ions which is bigger ionic radius Nd(0.983Å), Gd(0.938Å), Dy(0.912Å),

SC

In(0.8Å), and Y(0.9Å) than Mn(0.645Å) in Mn site increases the Mn4+ content by reducing the Mn3+ in the spinel structure which is Jahn-Teller active. The metal ion substitution suppresses the phase

M AN U

transition from spinel to tetragonal Li2Mn2O4, which is Jahn- Teller active, and improves the structural and electrochemical stability. Among the various doping elements, Sc3+ is one of the effective dopant for Mn3+ which will not induce Jahn-Teller distortion as there is no d orbital electron in Sc3+ ion. To our knowledge, only one report is available on Sc-doped LiMn2-xScxO4 where a preliminary

TE D

electrochemical study has been carried out at 0.1C rate for 50 cycles [26]. However, no further details available on structure, morphology and detailed electrochemical results. Herein, we report the detailed investigation on the synthesis of Sc-doped LiScxMn2-xO4 (LSMO)

EP

(where x = 0.00, 0.04, 0.06, 0.08 & 0.1) by solid state method to produce highly crystalline material. The influences of lattice parameter, particle morphology, size distribution and electrochemical

AC C

properties on Sc3+ doping in LMO crystal lattice were investigated. In addition, GITT, EIS and CV measurements were carried out to determine the Li-ion kinetics to understand the properties which leads to the improved electrochemical performance. Further we present ex-situ analysis, structure and morphology of cycled electrodes of LMO and LSMO to show the correlation between structure and electrochemical performances.

2 Experimental 3

ACCEPTED MANUSCRIPT

2.1

Materials and Methods Lithium hydroxide (LiOH.H2O), manganese acetate (MnCH3COO.4H2O), sodium per sulphate

RI PT

(Na2S2O8) were purchased from Merck and scandium oxide powder (Sc2O3) from Alfa Aeser and used as received. MnO2 nanorods were prepared by hydrothermal method [34] with a modified procedure as given in supporting information. Synthesis of LiScxMn2-xO4

SC

2.2

M AN U

LiScxMn2-xO4 was synthesized by solid state reaction. To a stoichiometric amount of LiOH.H2O (0.25g, 1mol) and MnO2 (0.96 g, 2 mol), an appropriate weight ratio of Sc2O3 (x = 0.0, 0.04, 0.06, 0.08 and 0.1) was added. The resulting mixture was thoroughly mixed using mortar and pestle in hexane medium for 1h. The black powder was calcinated in air at 800˚C for 12 h in a tubular furnace. The temperature ramping and cooling rate was maintained at 5˚C/min. Resulting materials were named as

LiSc0.1Mn1.9O4 as (5).

Physicochemical Characterization

EP

2.3

TE D

LiMn2O4 as (1), LiSc0.04Mn1.96O4 as (2), LiSc0.06Mn1.94O4 as (3), LiSc0.08Mn1.92O4 as (4) and

The XRD data were collected by Rigaku smart lab X-ray powder diffractometer (Cu-Kα,

AC C

wavelength λ= 1.5406 Å) with D/teX Ultra 250 detector. The XRD pattern was recorded in the 2θ range between 10° and 90° with a scan rate of 1°/min at 25°C. Rietveld refinement was performed with General Structure Analysis Software (GSAS/EXPUI) to calculate the lattice parameter ‘a’ and cell volume ‘V’. The morphology of the samples was observed by SEM, (Zeiss Merlin, LaB6 filament) equipped with energy dispersive X-ray spectroscopy (EDX). The powder samples were sprinkled on to a carbon tape and coated with a thin (Au:Pd 80:20) layer by a sputtering process to enhance the conductivity for SEM analysis. The ‘image-J’ software was used to find out the particles length and 4

ACCEPTED MANUSCRIPT

breath. The High Resolution Transmission Electron Microscopy (HR-TEM) imaging and selected area electron diffraction (SAED) images were taken by FEI, Tecnai G2HR-TEM with LaB6 filament and at an accelerating voltage of 200 kV. The powder samples for HR-TEM analysis were prepared by

RI PT

dispersing the samples in ethanol by a sonication process and these suspensions were drop casted on a 400 mesh holy carbon grid on copper support (Agar Scientific). The images and corresponding SAED pattern were obtained from random locations on sample to get on overview morphology of the samples.

SC

Infrared absorption spectra of the samples were recorded using a Fourier Transform Infrared Spectrometer (FTIR) with platinum ATR/FTIR technique (Bruker vacuum VERTEX 70 V). FTIR data

M AN U

were collected at a transmission mode with a spectral resolution of 2 cm−1 after 32 scans under vacuum. X-Ray photoelectron spectroscopy (XPS) analysis was obtained to analyze the surface chemistry (oxidation states and the elemental composition) of the materials using the ESCA-Omicron XPS

2.4

TE D

system with Mg-Ka as the excitation source.

Electrochemical Characterizations

Electrochemical properties of the compounds 1−5 as cathode materials against Li/Li+ were

EP

investigated by galvanostatic charge-discharge technique using a coin cell (CR2032) at room

AC C

temperature. The working electrode was fabricated by blending the slurry containing 80% active material, 15% Ketjan acetylene black (conducting carbon) and 5% polyvinylidene difluoride (binder) in N-methyl-2-pyrrolidone solvent. The slurry was coated on aluminium foil and dried at 120°C in hot air oven. The half-cell (LSMO vs. Li) was fabricated inside the argon filled glove box (O2 and H2O <1.0 ppm) using the coated electrode as working electrode, pure lithium foil as reference/counter electrode, layer of glass fiber (GF/D; Whatman) as a separator and 1M solution of LiPF6 in fluorinated-ethylene carbonate (FEC) /dimethyl carbonate (DMC, 1: 4 v/v) as an electrolyte. The Cyclic voltammetry (CV) measurements were carried out using a PARTSTAT MC electrochemical workstation at different scan 5

ACCEPTED MANUSCRIPT

rates in the potential range between 3.0 and 4.5 V vs. Li/Li+. The electrode active mass loading for all the samples were maintained in the range of 2.5−3 mg cm−2. Galvanostatic charge-discharge cycling

RI PT

was recorded at various current rate (1C = 148 mAg−1) in the potential range between 3.0 and 4.5 V using an Arbin BT-2000 (Arbin Instruments, USA). Here, 1C-rate corresponds to extraction or insertion of 1 Li from LiMn2O4 in one hour. Galvanostatic Intermittence Titration Technique (GITT) was carried out using Biologic instruments. Electrochemical impedance spectra (EIS) were measured

SC

by PARTSTAT MC electrochemical workstation in the frequency range from 1 MHz to 0.01 mHz at

M AN U

with AC signal (10 mV) amplitude. Electronic circuit model (Zview software) was used to fit the impedance spectra to calculate the resistance values during electrochemical performances. The cycled electrode were disassembled inside the glove box, washed with anhydrous DMC and dried, and further characterized by XRD and SEM analyses. Results and Discussion

TE D

3

The Sc-doped LiScxMn2-xO4 Spinels (where x = 0.00 (1), 0.04 (2), 0.06 (3), 0.08 (4) and 0.1 (5)) were synthesized by solid state reaction using MnO2 nanorods and LiOH as precursors. At first, MnO2

EP

precursor was synthesized by hydrothermal method and characterized by XRD and SEM. XRD pattern of as synthesized MnO2 powder confirms the tetragonal symmetry with P42/ mnm space and SEM

AC C

image with rod-like morphology (Supporting Information Fig. S1 and S2). XRD analysis was carried out to evaluate the structural information of LiMn2O4 and Sc-doped LiScxMn2-xO4 spinels. Fig. S3 (see Supporting Information) compares the XRD pattern of all the samples which could be indexed with Fd3m space group. The obtained diffraction patterns of all the samples are matched well with JCPDS file no: 98-006-2536 (Fd3m and SG: 227). Thus it confirms the formation of pure phase without formation of any secondary phase like LiScO2 or impurity like Sc2O3. A very small additional peak appeared at 2θ 31.6° corresponding to unreacted Sc2O3 when the doping concentration exceeded above 0.04. The 6

ACCEPTED MANUSCRIPT

lattice parameter changes on Sc-doping were calculated by LeBail Rietveld refinement method on the basis of LiMn2O4 crystal structure with Fd-3m space group. The observed X-ray diffraction pattern, the calculated and their difference pattern of all the samples are shown in the Fig. 1(a-e) The good

RI PT

agreement between the calculated pattern and the observed pattern is indicative of a good refinement. The refined pattern shows sharp peaks suggesting that the samples are in crystalline form. The agreement factor, χ2, lattice parameter, ‘a’ cell volume ‘V’ interatomic distance of all the samples were

SC

obtained from the refinement analysis and the results are tabulated (Table 1). Fig. 1f shows the plot of Sc concentration vs. lattice parameter. The value of lattice parameter ‘a’ is decreases as the Sc content

M AN U

increases from 0 to 0.1. This is mainly due to higher binding energy of Sc−O (674 KJmol−1) than Mn−O (402 KJ mol−1) and increased Mn4+ content (less ionic radius of Mn4+, 0.53Å than that of Mn3+, 0.65 Å) in the spinel lattice upon Sc-doping. Thus the decrease in the lattice parameter confirms the incorporation of Sc-ion into the spinel lattice. Similar observation was reported in the literature for

TE D

doping of metal ions into LiMn2O4 lattice [29, 35, 36]. The increased bond length of Li−O in LiO4 tetrahedron facilitates the fast lithium ion diffusion in the matrix. In addition, the decreased Mn−O bond length in MnO6 octahedraon framework enhances the structural stability resulting in less volume

EP

change during charge/discharge cycling.

AC C

The surface morphologies of 1−5 were analysed by SEM, the SEM images of 1, 3 and 5 are shown in Fig. 2 (for 2 and 4 see Supporting Information Fig. S4). All compounds show faceted rod-like morphology. The length of rods in sample 1 was found to be in the range of 50-200 nm, whereas, the rod length in 2, 3, 4 and 5 are in the range of 100-600 nm. The length of the rod increases with increase in Sc content; this confirms that the doping amount changes the particle size of the samples significantly. The aspect ratio (length/breath) and frequency count profile is shown in the Fig. 2.

7

ACCEPTED MANUSCRIPT

Among various compositions, 3 show a narrow aspect ratio with an average value of 1.3. Local Energy dispersive X-ray (EDX) confirms the presences of Sc and Mn in the matrix. HRTEM images of 3 show well-defined crystals with the size range between 200-400 nm

RI PT

(Fig.3a), which is corroborating well with SEM results. The selected area electron diffraction of 3 performed at various random locations showed a similar pattern as in Fig. 3b. From the lattice fringes of HRTEM images, interplanar ‘d’ spacing corresponds to the (111) lattice planes of cubic 3 was

SC

determined to be 0.47 nm which is characteristic of cubic spinel lattice (JCPDS:98-006-2536). The value of interplanar distance is in good agreement with the measured d-spacing value of 0.47 nm for

M AN U

(111) hkl plane determined by XRD data using Bragg’s law.

Infrared transmittance spectrum of 1 (Fig. S5) show two peaks at 616 and 600 cm−1 for the symmetric stretching vibrations of Mn−O bond in the octahedral MnO6. FT-IR spectrum of 3 shows the symmetric stretching vibration Mn−O bond at 613 and 596 cm−1. These bands shifted to lower wave

TE D

numbers (~4 cm−1) which can be ascribed to the occupancy of Sc3+ ion in the octahedral site. The similar observation was also reported for Dy3+ doped spinel structure [37]. In addition, three weak bands were observed at ~395, 340, and 261 cm−1. The band at 395 cm−1 is attributed to the asymmetric

EP

stretching vibration of Li−O bond of LiO4 tetrahedra in the spinel structure and other two bands

AC C

correspond to the bending modes of O−Mn−O bonds [38]. To analyse the chemical environment of the elements, XPS analysis was carried out for 1 and 3. The XPS survey spectrum of 1 and 3 shows C, O, Mn, Li peaks and in addition Sc peak was observed in 3 (see Supporting Information Fig. S6 (a,b)). The spectra of C1s and O1s for both the samples are given in Fig. S6 (c-f). The C1s contains three peaks at 284.8, 286.1 and 288.8 eV correspond to C−C, C−O−C and O−C=O bonds, respectively. The C−C peak at 284.8 eV was used for calibration. The O1s spectrum of 3 is dominated by metal oxide peak at 530.2 eV. It also contains C−O species of Li2CO3 8

ACCEPTED MANUSCRIPT

which is commonly found in lithiated metal oxides [39]. The spectra of Li1s, Mn3p and Sc2p region are shown separately in the Fig. 4(a,b,d). The Li 1s peak observed for 1 at 53.7 eV and for 3 at 54 eV. In the sample 1, a broad band was observed at 49.84 eV for Mn 3p3/2 and Mn 3p1/2 , this might be due

RI PT

to the presence of equal content of Mn4+ and Mn3+ ions. In the case of 3, a peak at 50.1 eV and a shoulder at around 49.1 eV were appeared, which corresponds to Mn 3p3/2 and Mn 3p1/2 splitting. The obtained peak positions of Mn4+ and Mn3+ in Mn 3p spectra are comparable with the reported values

SC

[39]. The observed peak intensity of the Mn4+ is relatively higher than the Mn3+ confirms the reduction of Mn3+ content upon Sc3+ doping. Two major peaks correspond to Mn 2p3/2 and Mn 2p1/2 was observed

M AN U

at 641.8 and 653.4 eV for LiMn2O4. In comparison, for doped spinel these peaks were shifted to higher values (642.8 and 654.4 eV) (Fig 4b). The increased values of Mn2p and Li in 3 indicating that change in the lattice environment compared to 1. In addition, two distinct peaks observed at 402.5 eV and 407.2 eV in 3 are corresponding to Sc 2p3/2 and Sc 2p1/2. Whereas, the peak positions of Sc 2P3/2 and Sc

TE D

2p1/2 in Sc2O3 powder reported to be 401.71 and 406.46 eV, respectively [40]. The observed shift in the binding energy indicates that the environment around Sc3+ in 3 is different from that of Sc2O3 leading to the conclusion that Sc3+ is doped in the spinel lattice. Mn2P3/2 signal is very complex with multiple

EP

splitting of the main peak. Fig. 4c shows the Mn2P3/2 spectra of compound 3 with curve fitting. The valance state of Mn and percentage of particular valance were calculated by curve fitting method. From

AC C

the fitting results, the percentage area of Mn4+ (56.5%) is found to be higher than Mn3+ (43.5%). The deviation of Mn3+ (6.5%) concentration is an additional evidence for Sc-doping into the spinel. The increased Mn4+ state can be correlated with the lattice parameter values of XRD. Lattice values were reduced due to presence of higher amount Mn4+ whose ionic radius (0.53Å) was lesser than Mn3+ (0.65Å).

9

ACCEPTED MANUSCRIPT

Cyclic Voltammograms (CVs) of the compounds 1-5 exhibited similar profiles in the potential range of 3.5−4.5 V vs. Li/Li+ (see Supporting Information Fig. S7 for 1 and 3). CV profiles of 1 exhibit

RI PT

two oxidation (4.06 and 4.18V) and two reduction (4.06 and 3.93V) peaks for Mn3+/Mn4+ redox couple. Sample 3 exhibits similar oxidation (4.10 and 4.22 V) and reduction (4.03 and 3.91V) peaks. The first redox couple can be attributed to the removal/insertion of Li+ from half of the tetrahedral sites in which Li−Li interaction exists and the second redox couple corresponds to the removal/insertion of Li+ from

M AN U

Li0.5Mn2O4 and forms Li0.25Mn2O4 finally [26, 41].

SC

the other site where no Li−Li interactions present, thus involves phase transition of LiMn2O4 to

Fig. 5a shows the first charge/discharge profile of 1-5 at 0.1C rate and the corresponding dQ/dV plots are given in the Supporting Information Fig.S7. Sample 1 exhibited two plateaus for oxidation at 4.00 and 4.14 V and the corresponding reduction at 4.10 and 3.99 V(based on dQ/dV plot). Whereas,

TE D

the oxidation plateaus of the doped samples showed in the potential range of 4.00 - 4.03V and 4.134.14 V and reduction plateaus appeared in between 4.10 - 4.12V and 3.98- 3.99 V, which are consistent with the reported literature values [32, 42−47]. Compound 1 delivered an initial charge

EP

capacity of 140 mAhg−1and the corresponding discharge capacity of 126 mAhg−1(theoretical capacity of 148 mAhg−1) with a columbic efficiency of 90%. The compounds 2-5 exhibited the initial

AC C

charge/discharge capacities of 128/119, 122/119, 119/115 and 112/111mAhg−1 with columbic efficiency of 93%, 95%, 96% and 99%, respectively. On Sc-substitution, the first charge capacity was decreased, however, the first cycle irreversible capacity loss decreased significantly with increasing Scdoping. Subsequently after two charge/discharge cycles at 0.1C rate, the cells were tested at 1C rate for 100 cycles. At 1C rate, 1-5 delivered charge/ discharge capacities of 121/117, 112/106, 116/114, 110/107 and 110/108 mAhg−1, and after 100 cycles they showed charge/discharge capacities of 98/97, 104/103, 111/110, 106/106 and 106/105 mAhg−1, with capacity retention of 85%, 97%, 96%, 95% and 10

ACCEPTED MANUSCRIPT

97%, respectively. It’s evident that Sc- substitution enhances the cycling stability of the spinel. Among all the compounds, 3 exhibited excellent cycling stability (Fig. 5b), indicating that LiSc0.06Mn1.94O4 composition is the optimum Sc-substitution in the spinel structure. Further long cycle life test was

RI PT

carried out for both 1 and 3 at 1C rate (see Fig. 5c). At the end of 500 cycles, 3 showed the charge/discharge capacity of 104/102 mAhg−1 with capacity retention of 89%, whereas, 1 delivered charge/discharge capacity of 90/85 mAhg−1 with 74% retention. The 26% capacity fade was observed

SC

for 1 is due to (i) the presence of higher content of Mn3+ in LiMn2O4 spinel will undergo

M AN U

disproportionation reaction to 2Mn3+ → Mn4+ +Mn2+. Due to this, Mn2+ (0.83 Å) will dissolves into the solution and deposit on both side of the electrode by reducing into Mn in the electrolyte. Manganese deposition on the both of the electrode surface further deteriorates the lithium ion kinetics and increases the charge transfer resistance as confirmed by impedance analysis. (ii) The Mn3+ ion promotes JahnTeller distortion by volume expansion upon charging/discharging and formation of secondary

TE D

tetragonal phase Li2Mn2O4 leads to structural instability. Sample 3 shows better stability than 1 due to decrease in dissolution of Mn2+ ion during charge/discharge process which is mainly due to the decreased Mn3+ and increased bond strength of MnO6 octahedron framework by reducing the Mn4+−O

EP

bond length upon Sc3+ ion doping. The contracted MnO6 octahedron in the spinel structure helps to

AC C

reduce the volume change upon cycling and improves the cycling stability. The rate capability tests were also performed for 1 and 3 by varying the current rate from 0.1C to 15C as shown in the Fig.5d. At first, the cells were cycled at 0.1C for 10 cycles, and then the C-rate was increased stepwise up to 15C, for each current rate 10 charge/discharge cycles were conducted. A summary of charge/discharge values at end of the 10 cycles at each C-rate were provided in the Table 2. Compound 3 delivered a discharge capacity of 116 mAhg−1 at 0.1C which is less than first discharge capacity of compound 1. This is due to the reduction of Mn3+ content in this spinel by doping 11

ACCEPTED MANUSCRIPT

electrochemically inactive Sc3+ ion. It is apparent that the discharge capacities decrease upon increasing the current rate (0.1C to 15C). Even though 1 delivered a high discharge capacity of 126 mAhg−1 at 0.1 C, at the end of 10 cycles the capacity decreases to 117 mAhg−1 with capacity retention of 92%. On the

RI PT

other hand, 3 delivered an initial capacity of 116 mAhg−1 at 0.1C with excellent capacity retention of 98% after 10 cycles. As C rate increases compound 3 exhibited better discharge capacities than 1. Moreover, compound 1 showed drastic fade in capacity after 5C rate, while compound 3 exhibited a

SC

reasonable capacity of 42 mAhg−1 at 15C. After 170 cycles, the C-rate reduced to 0.1C, 3 delivered a

M AN U

capacity of 110 mAhg−1 with capacity retention of >99%. It is suggesting that compound 3 exhibited better rate capability than 1, which can be ascribed to the fast lithium ion kinetics in the spinel structure by expanding the LiO4 tetrahedron due to increased Li−O bond length upon Sc3+ doping. The doped Sc3+ ions partially occupy at 16d sites without changing the crystal structure of the spinel. The doping of Sc3+ in place of Mn3+ in LiMn2O4 in the octahedral framework reduces the Jahn-Teller distortion as

TE D

well as Mn dissolution. Furthermore Sc−O bonding provides an improved structural stability to this spinel due to its higher binding energy (647 KJ/mol) than Mn−O bond (402 KJ/mol). As a result better

EP

capacity retention and cyclic stability was observed for Sc-doped spinel than the undoped LMO. EIS analyses were carried out for both 1 and 3 after the rate capability experiment. Fig. 6a

AC C

shows the Nyquist plots (Z’ vs –Z”) of the samples. The impedance spectra of both 1 and 3 consist of depressed semicircle in the high frequency range and a straight line at medium frequency range. The depressed semicircle is a convolution of two separate semicircles, which is confirmed by fitting of the impedance data with an equivalent circuit using Zview software. The first semicircle at the high frequency region can be ascribed to the migration of Li ion through solid electrolyte interface (SEI) layer. The second semicircle in the middle frequency region is corresponding to the charge transfer resistance between particles and electrolyte. The impedance data was fitted with two RC equivalent 12

ACCEPTED MANUSCRIPT

circuits as shown in Fig. 6a (insert). The circuit elements were deduced by fitting the equivalent circuit is solution resistance (Rs), capacitance (CPE1 and CPE2), surface film resistance (RSEI), charge transfer resistance (Rct) and Warburg resistance (RW). The obtained values of Rs = 8 Ω, RSEI = 63 Ω and Rct =

RI PT

146 Ω for 1 and Rs = 4.9 Ω, RSEI = 94 Ω, Rct = 26 Ω for 3 suggesting that compound 1 has significantly higher charge transfer resistance than 3. The increase of charge transfer resistance could be due to increased Mn2+ ion dissolution into the electrolyte during prolonged cycling at high current and

SC

deposition Mn on the electrode surface. The deposition of Mn on the spinel electrode during charging/discharging hinders the lithium ion kinetics and increases the overall resistance.

M AN U

The rate performance of the composite electrode depends on the diffusion capability of the Li+ ion into the host materials and its electronic conductivity as well as particle size. The diffusion capability is proportional to the diffusion coefficient (DLi+) and inversely proportional to the square of the particle size (diffusion capability = D/r2) [48]. The DLi+ of electrodes can be determined by GITT

TE D

[17, 49, 50] CV [51] and EIS [52] methods. We have determined the DLi+ for 1 and 3 by GITT, CV and EIS techniques. For GITT method, at first the cells were charged/ discharged at 0.05C rate from 3.0 – 4.5 V for two cycles. Subsequently the cells were kept under equilibrium (E0) for one day. Then the

EP

cells were charged at 0.1 C rates for 15 min (τ), followed by open circuit relaxation for 60 min to reach the steady state voltage (Es). This process was repeated for the entire voltage window between 3.0 and

AC C

4.3V. At the end of charging, the compounds have known amount Li contents, x+∆x depending on the applied current. The voltage vs. time profile in GITT is depicted in Fig. 7a. While charging, the equilibrium cell voltage (E0) increases with time and is superimposed on IR drop due to the current flux through the electrolyte and interface. A change in steady state voltage ∆Eτ = (Eτ – E0) during the charging process can be calculated by subtracting the equilibrium potential (E0) and IR drop from the steady state voltage (Eτ). A change in steady state value ∆Es = (Es–E0) was calculated from single

13

ACCEPTED MANUSCRIPT

titration profile for entire voltage vs. time profile. The DLi+ values were calculated using the modified Fick’s second law of diffusion (Equation 1) [51].  





 

 

 





…. (1)

RI PT


 = 



where, mB is the active material mass in the electrode (g mol−1), Vm is molar volume (cm3 mol−1 and MB is formula mass of the electrode (g cm−3). A is geometrical area of the electrode (1.77 cm2). ∆Es is

SC

change in steady state potential and ∆Eτ is the total change in cell voltage and τ is relaxation time. Fig. 7b displays the single titration curve of 3 labelling with different parameters Es, Eτ and E0

M AN U

etc. The voltage vs. τ1/2 profile exhibits straight line behaviour (Fig. 7c). The DLi+ calculated from GITT curve of 1 and 3 as a function of cell voltage is shown in the Fig. 7d. The calculated DLi+ values for 1 and 3 during charging lie in the range of 1×10−10-10−11cm2 s−1. During discharging the DLi+ values of 1 are in the range of 1×10−10- 10−13 cm2 s−1, whereas for 3 the values lie in the range of 1 ×10−10 - 10−12

TE D

cm2 s−1 [51, 53, 54]. It is noteworthy to mention that the diffusion co-efficient value of 3 is one order higher than that of 1. The intensity minimum at 4.12V in the DLi+ vs. voltage profile of 3 decreased compared to that of 1 suggesting the marginal suppression of the transition from one cubic to another

EP

cubic phase during charging/discharging. Hence, the inter particle contact is enhanced and the volume

AC C

change decreased for 3, resulting in increased Li+ ion diffusion and electrical conductivity. Further, DLi+ values of 1 and 3 determined by CV and EIS techniques are in good agreement with GITT technique (see Supporting Information Fig. S8). In order to check the morphological and structural information of the prolonged cycled electrodes, SEM and XRD analyses were carried out for 1 and 3 after 500 charge/discharge cycles at 1C rate. Fig. 8(a-h) shows SEM and XRD measurements of the electrodes 1 and 3 ( Fig. S9 Supporting Information). The SEM images of LiMn2O4 electrode before and after electrochemical cycling show 14

ACCEPTED MANUSCRIPT

dark spots which are rich in Mn. The dark spots are seen in the pristine electrode of the Sc-doped phase but, to a lesser extent. However, no dark spots are evident if the electrode made of the Sc-doped phase after electrochemical cycling. In addition, the electrode of the undoped phase shows severe cracks

RI PT

being present; the cracking is much less in the electrode made of the Sc-doped phase (Fig.8e,g) The absence of the Mn-rich regions and negligible cracking of the electrode in the case of the Sc-doped phase suggests that the integrity of the electrode is well maintained. This can be an indirect evidence

SC

for the mitigation of dissolution on Mn from the electrode in the Sc-doped phase vis a vis the undoped phase.

M AN U

The main XRD peaks exhibited similar diffraction patterns as those of 1 and 3 shown in Fig. 1. However a well-pronounced peak broadening along with peak shift to lower angle was observed in the case of 1. It can be seen from Fig. 8(h) other peaks are appeared at 19.2, 37.1, and 43.6 in 2θ for 1, which could be due to the formation of λ–MnO2 phase. Whereas, the spectrum of 3 shows neither

TE D

significant peak broadening nor impurity peaks. The peak broadening is related to structural disorder in the spinel which could be formed as a result of the dissolution of Mn2+ ions. The lattice parameter was calculated for both 1 and 3 after charge/discharge cycling. The cell parameter of 1 was contracted from

EP

8.232 to 8.134 Å (0.098 Å). On the other hand the lattice parameter contraction for sample 3 from 8.229 to 8.221Å (0.008 Å), which is much smaller than 1. It is well-known phenomenon that the

AC C

contraction of cell parameter in the cycled LiMn2O4 is due to Mn2+ dissolution. The loss of Mn ion from the spinel structure and contraction of the other atomic bonds are the main reasons for the reduced lattice parameter.

4

Conclusion A series of Sc-doped spinels LiScxMn2-xO4 (x = 0.04, 0.06, 0.08 and 0.1) have been synthesized

successfully by solid state reaction. All the compounds exhibited face-centered cubic structure with an 15

ACCEPTED MANUSCRIPT

Fd-3m space group. Sc-doping considerably reduces the lattice parameters and bond lengths of the spinel structure by changing the inter-atomic distances. The doped spinels show rod-like morphology

RI PT

with an average aspect ratio of 1.3. LiSc0.06Mn1.94O4 delivers a discharge capacity of 114 mAh g−1 at 1C with excellent cyclic stability over 500 cycles at capacity retention of >90%. In addition, LiSc0.06Mn1.94O4 exhibit good rate capability due to the low charge transfers resistance and enhanced lithium-ion diffusion kinetics. Moreover, the structural and morphological analyses of the cycled

SC

electrode confirms that in Sc-doped spinel structure remains intact without any segregation of Mn-rich agglomerates even after 500 charge/discharge cycles. The excellent electrochemical performances of

M AN U

the doped spinel can be attributed to the structural stability derived from Sc-doping. It is suggested that Sc-doped spinel could be a promising cathode material for lithium ion battery for EV applications.

Acknowledgements

TE D

This work was financially supported by ARCI –Technical Research Centre [Ref No.AI/1/65/ARCI/2014 (c)] through Department of Science and Technology (DST), India is highly acknowledged. The authors are thankful to G. Padmanabham, Director ARCI and Prof. G.

EP

Sundararajan, Distinguished Emeritus Scientist, ARCI. Authors are thankful to Dr. S. Anandan for XPS, Dr. B. V. Sarada for Raman Dr. D. Prabhu (Scientist, ARCI) for SEM and Mr. Ravi Gautam

AC C

(Junior Scientist, ARCI) for HR-TEM measurements.

References

1. Bruce Dunn, Haresh Kamath, Jean-Marie Tarascon, 2011, 334, 928-935. 2. V. Etacheri, R. Marom, R. Elazari, G. Salitra, D. Aurbach, Energy Environ., Sci., 2011, 4 3243–326 . 3. R. Prakash, A.K. Mishra, A. Roth, C. Kubel, T. Scherer, M. Ghafari, H. Hahn, M. Fichtner, J. 16

ACCEPTED MANUSCRIPT

Mater. Chem. 20 (2010) 1871–1876. 4.

R. Prakash, C. Wall, A. K.Mishra, C. Kubel, M. Ghafari, H. Hahn, M. Fichtner J. Power Sources 196 (2011) 5936–5944.

RI PT

5. G. Fang, J. Zhou, Y. Cai, S. Liu, X. Tan, A. Pan, S. Liang, J. Mater. Chem. A, 5(2017) 1398313993.

6. T. Chen, J. Zhou, G. Fang, Y. Tang, X. Tan, A. Pan, S. Liang, ACS Sustainable Chem. Eng.

SC

2018, 6, 7250−7256.

M AN U

7. M.M Thackeray, P.J Johnson, L.A. de Picciotto, P.G Bruce, J.B.Good enough, Mat.res.Bull., 1984, 19, 179-187.o 8. K. Nishimura, T. Douzono, M. Kasai, H. Andou, Y. Muranaka, Y. Kozono, J. Power Sources 1999, 81–82 420–424.

9. J.S. Kim, K. Kim, W. Cho, W.H. Shin, R. Kanno, J.W. Choi, Nano Lett. 12 (2012) 6358–6365.

TE D

10. J.M. Tarascon, E. Wang, F. K. Shokoohi, J. Electrochem. Soc. 138 (1991) 2859-2864. 11. S. Lee, G. Yoon, M. Jeong, M.J. Lee, K. Kang, J. Cho, Angew. Chemie - Int. Ed. 54 (2015) 1153–1158.

EP

12. K.Y. Chung, C.W. Ryu, K.B. Kim, J. Electrochem. Soc. 152 (2005) A791-A795. 13. M.M. Thackeray, Y. Shao-Horn, A.J. Kahaian, K.D. Kepler, E. Skinner, J.T. Vaughey, S.

AC C

Hackney, Electrochem. Solid-State Lett. 1 (1998) 7–9. 14. Y. Dai, L. Cai, R. E. White, J. Electrochem. Soc. 160 (2013) A182–A190. 15. J. Dong H, Y.J. Shin, S.M. Oh, J. Electrochem. Soc. 143 (1996) 2204–2211. 16. A. Yamada, J. solid state chemistry 122 (1996) 160-165. 17. M. Hirayama, H. Ido, K. Kim, W. Cho, J. Am. Chem. Soc. 132 (2010) 15268–15276. 18. C. Yang, Y. Deng, M. Gao, X. Yang, X. Qin, G. Chen, Electrochim. Acta. 225 (2017) 198– 206. 17

ACCEPTED MANUSCRIPT

19. Y.K. Sun, K.J. Hong, J. Prakash, J. Electrochem. Soc. 150 (2003) A970–A972. 20. J. S. Gnanaraj, V. G. Pol, A. Gedanken, D. Aurbach, Electrochem. Commun. 5 (2003) 940–945.

22. Y. Gao, J. R. Dhan, J. Electrochem.soc. 143 (1996) 100-114.

RI PT

21. D. Guan, C. Cai, Y. Wang, 2011 IEEE Green Technol. Conf. Green 2011 1465–1469.

23. Y.K. Lee, J. Park, W. Lu, J. Electrochem. Soc. 163 (2016) A1359–A1368.

24. Y.L. Ding, J. Xie, G.S. Cao, T.J. Zhu, H.M. Yu, X.B. Zhao, J. Phys. Chem. C 115 (2011)

SC

9821–9825.

25. I. Taniguchi, Z. Bakenov, Powder Technol. 159 (2005) 55–62.

M AN U

26. F. Zhou, X. Zhao, H. Zheng, Z. Zhang, M. Ji, Mater. Lett. 58 (2004) 3720–3724. 27. Singhal S.R. Das, M.S. Tomar, O. Ovideo, S. Nieto, R.E. Melgarejo, R.S. Katiyar, J. Power Sources 164 (2007) 857–861.

28. S.C. Han, S.P. Singh, Y. Hwang, E.G. Bae, B.K. Park, K. S. Sohn, M. Pyoa, J. Electrochem.

TE D

Soc. 159( 2012) A1867-A1873.

29. P. Rama, R. Singhal, G. Choudharya R.K. Sharmaa, J. Electro analytical Chemistry 802 (2017) 94–99.

EP

30. H.W Liu, L.Q. Feng, H. Tang, L. Song, K.L. Zhang, J. material science: materials in electronics

AC C

15 (2004) 495-497.

31. X. He, J. Li, Y. Cai, Y. Wang, J. Ying, C. Jiang, C. Wan, J. of Power Sources, 2005, 150, 216– 222.

32. M.R. Palacı́n, G.G. Amatucci, M. Anne, Y. Chabre, L. Seguin, P. Strobel, J. M. Tarascon, G. Vaughan, J. Power Sources 81-82 (1999) 627–631. 33. Q. Jiang, D. Liu, H. Zhang, S. Wang, J. Phys. Chem. C 119 (2015) 28776–28782. 34. D.K. Kim, P. Muralidharan, H.W. Lee, R. Ruffo, Y. Yang, C.K. Chan, H. Peng, R.A. Huggins, 18

ACCEPTED MANUSCRIPT

Y. Cui, Nano Lett. 8 (2008) 3948–3952. 35. Y.J. Wei, L.Y. Yan, C.Z. Wang, X.G. Xu, F. Wu, G. Chen, J. Phys. Chem. B, 108 (2004) 8547– 18551.

RI PT

36. Y.K. Sun, C. S. Yoon, C. K. Kim, S. G. Youn, Y.-S. Lee, M. Yoshio and I.-H. Oh, J. Mater. Chem., 2001, 11, 2519–2522.

37. S. Balaji, T. Mani Chandran and D. Mutharasu, Ionics, 2012, 18, 549–558.

SC

38. C. Julien and M. Massot, Phys. Chem. Chem. Phys., 2002, 4, 4226–4235.

39. S. R. Kasireddy, B. Gangaja, S. V. Nair and D. Santhanagopalan, Electrochim. Acta., 2017,

M AN U

250, 359–367.

40. M. C. Biesinger, L.W. M. Lau, A. R. Gerson, R. St. C. Smart, Applied Surface Science 2010, 257, 887–898

41. X. Zhao, M. V. Reddy, H. Liu, S. Ramakrishna, G.V.S. Rao and B.V.R. Chowdari, RSC Adv.,

TE D

2012, 2, 7462–7469.

42. P. Ragupathya, H.N. Vasana and N. Munichandraiah, Materials Chemistry and Physics, 2010, 124, 870–875.

EP

43. W. Tang, X. Yang, Z. Liu, S. Kasaishi and K. Ooi, J. Mater. Chem., 2002, 12, 2991–2997. 44. X. Xie, D. Su, B. Sun, J. Zhang, C. Wang and G. Wang, Chem. -A Eur. J, 2014, 20, 17125–

AC C

17131.

45. M. Xiang, C. W. Su, L. Feng, M. Yuan and J. Guo, Electrochim. Acta., 2014, 125, 524–529. 46. M. Helan, L.J. Berchmans and A.Z. Hussain, Ionics, 2010, 16, 227–231. 47. M. V Reddy, S. S. Manoharan, J. John, B. Singh, G.V.S. Rao and B.V.R. Chowdari, J. Electrochem. Soc.,2009, 156, A652–A660. 48. Yang, B. Yan, J. Ye, X. Li, Y. Liu and H. You, Phys. Chem. Chem. Phys., 2014, 16, 2882 – 2891. 19

ACCEPTED MANUSCRIPT

49. W. Weppner, J. Electrochem. Soc., 1977, 124, 1569-1578. 50. K. M. Shaju, G. V. Subba Rao and B.V.R. Chowdari, J. Mater. Chem., 2003, 13, 106-113.

Electrochem. Soc., 2005, 152, A1283–A1289.

RI PT

51. J. R. Dahn, J. Jiang, L. M. Moshurchak, M. D. Fleischauer, C. Buhrmester and L. J. Krause, J.

52. T.-F. Yi, J.-Z. Wu, J.-Yuan, Y.-R. Zhu, and P.-F. Wang, ACS Sustainable Chem. Eng., 2015, 3, 3062−3069.

SC

53. M. Wakihara, L. Guohua, H. Ikuta and T. Uchida, Solid State Ionics, 1996, 86-88, 907-909. 54. D. G. Fauteux, A. Massucco, J. Shi and C. –O. Nnerud, J. Appl. Electrochem., 1997, 27, 543-

AC C

EP

TE D

M AN U

549.

20

ACCEPTED MANUSCRIPT Table list: 1. Table 1. Goodness of fit χ2 and refined lattice Parameter ‘a’, cell volume ‘V’ and bond length of Li−O and Mn−O of LiMn2O4, LiSc0.04Mn1.96O4, LiSc0.06Mn1.94O4, LiSc0.08Mn1.92O4 and LiSc0.1Mn1.9O4. 2. Table 2. Initial discharge capacity and capacity retention (at the end of the 10th cycles)

RI PT

of LMO (1) and LSMO (3) at different C rates.

Lattice Unit cell parameter volume‘V’ ‘a’ (Å) (Å)3

Bond length

Bond length

M AN U

χ2 Value

Compound Name

SC

Table 1. Goodness of fit χ2 and refined lattice Parameter ‘a’, cell volume ‘V’ and bond length of Li−O and Mn−O of LiMn2O4, LiSc0.04Mn1.96O4, LiSc0.06Mn1.94O4, LiSc0.08Mn1.92O4 and LiSc0.1Mn1.9O4

Li− −O (Å)

Mn− −O(Å)

8.232

558.022

1.9578

1.9622

LiSc0.04Mn1.96O4 1.594

8.230

557.65

1.9504

1.9654

LiSc0.06Mn1.94O4 1.811

8.229

557.476

1.9676

1.9562

LiSc0.08Mn1.92O4 1.786

8.227

557.114

1.9689

1.9549

1.962

8.225

556.574

1.9627

1.9574

EP

LiSc0.1Mn1.9O4

TE D

1.975

LiMn2O4

AC C

Table 2. Initial discharge capacity and capacity retention (at the end of the 10th cycles) of LMO (1) and LSMO (3) at different C rates. Initial discharge capacity [mAh/g] (capacity retention after 10 cycles )

Compound

0.1C

0.5C

1C

2C

3C

5C

8C

10C

LMO (1)

126 (92%)

111 (97%)

103 (99%)

98 (99%)

93 (99%)

87 (99%)

42 (92%)

20 (99%)

LSMO (3)

116 (98%)

111 (99%)

111 (99%)

108 (99%)

108 (99%)

103 (99%)

99 (99%)

96 (99%)

12C

15C

0.1(re)

-

73 (96%)

-

108 (95%)

42 (99%)

110 (99%)

ACCEPTED MANUSCRIPT Figure List: Fig. 1 Refined XRD results show observed, calculated and difference patterns of (a) LiMn2O4(1), (b) LiSc0.04Mn1.96O4(2), (c) LiSc0.06Mn1.94O4(3), (d) LiSc0.08Mn1.92O4(4), (e) LiSc0.1Mn1.9O4 (5), and (f) Lattice parameter vs. Scandium content profile.

2.

Fig. 2 Compares the SEM images of (a) 1 (b) 3 (c) 5 and their corresponding particle size distribution.

3.

Fig. 3 (a) TEM images of LiSc0.06Mn1.94O4 particles and insert shows the diffraction pattern of (111), (311) and (110) planes and (b) corresponding SAED pattern.

4.

Fig. 4 (a) Li 1s/Mn 3p and (b) Mn 2p spectra of LiMn2O4 and LiSc0.06Mn1.94O4, (c) Mn2p3/2 fitting spectrum and (d) Sc 2p spectrum of LiSc0.06Mn1.94O4 .

5.

Fig. 5 Comparison of (a) charge/discharge profile of first cycle of 1-5 in the voltage range between 3.0 V and 4.5 V vs. Li at the current rate of 0.1C (b) Cycling stability of 1-5 (100 cycles at 1C current rate) (c) discharge capacity vs. cycle number of 1 and 3 for 500 cycles (d) rate capability profile of 1and 3.

6.

Fig. 6 (a) Impedance spectra of 1 and 3 after rate capability test and insert shows the equivalent circuit used to evaluate the impedance spectra using Zview program (b) graph of impedance real part (Zre) vs. angular frequency (ω−1/2).

7.

Fig. 7 (a) GITT charge/discharge curve of LiMn2O4 and LiSc0.06Mn1.94O4 (b) applied current flux vs. voltage profile of single titration at charge voltage of 4.11 V (c) voltage changes as a function of square root of time (τ1/2) during a single titration process at 4.11 V and (d) Li+ diffusion coefficient values calculated for LiMn2O4 and LiSc0.06Mn1.94O4 as a function of voltage.

8.

Fig. 8 (a,b) and (c,d) SEM image and corresponding EDX spectra of 1 and 3 electrodes before cycling, (e, f) SEM image and EDX spectrum of 1 after 500 charge/discharge cycles, (d) SEM image of 3 after 500 charge/discharge cycles and (h) XRD pattern of 1 (A) and 3 (B) of cycled electrode (after 500 cycles).

AC C

EP

TE D

M AN U

SC

RI PT

1.

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

Fig. 1 Refined XRD results show observed, calculated and difference patterns of (a) LiMn2O4 (1),(b) LiSc0.04Mn1.96O4(2), (c) LiSc0.06Mn1.94O4(3), (d) LiSc0.08Mn1.92O4(4), (e) LiSc0.1Mn1.9O4(5), and (f) Lattice parameter vs. Scandium content profile.

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

Fig. 2 Compares the SEM images of (a) 1 (b) 3 (c) 5 and their corresponding particle size distribution.

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

Fig. 3 (a) TEM images of LiSc0.06Mn1.94O4 particles and insert shows the diffraction pattern of (111), (311) and (110) planes and (b) corresponding SAED pattern.

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

Fig. 4 (a) Li1s/Mn3p and (b) Mn 2p spectra of LiMn2O4 and LiSc0.06Mn1.94O4 (c) Mn2p3/2 fitting spectrum and (d) Sc 2p spectrum of LiSc0.06Mn1.94O4 .

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

Fig. 5 Comparison of (a) charge/discharge profile of first cycle of 1-5 in the voltage range between 3.0 V and 4.5 V vs. Li at the current rate of 0.1C (b) Cycling stability of 1-5 (100 cycles at 1C current rate) (c) discharge capacity vs. cycle number of 1 and 3 for 500 cycles (d) rate capability profile of 1and 3.

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

Fig. 6 (a) Impedance spectra of 1 and 3 after rate capability test and insert shows the equivalent circuit used to evaluate the impedance spectra using Zview program (b) graph of impedance real part (Zre) vs. angular frequency (ω−1/2).

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

Fig. 7 (a) GITT charge/discharge curve of LiMn2O4 and LiSc0.06Mn1.94O4 (b) applied current flux vs. voltage profile of single titration at charge voltage of 4.11 V (c) voltage changes as a function of square root of time (τ1/2) during a single titration process at 4.11 V and (d) Li+ diffusion coefficient values calculated for LiMn2O4 and LiSc0.06Mn1.94O4 as a function of voltage.

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

Fig. 8 (a, b) and (c, d) SEM image and corresponding EDX spectra of 1 and 3 electrodes before cycling (e, f) SEM image and EDX spectrum of 1 after 500 charge/discharge cycles (d) SEM image of 3 after 500 charge/discharge cycles and (h) XRD pattern of 1 (A) and 3 (B) of cycled electrode (after 500 cycles).

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT