Received 2 November 2015; accepted 12 January 2016; published 15 January 2016
Spinel LiMn2O4 has been zeroed in attractive and promising cathode materials for lithium-ion batteries owing to its high voltages, proper Mn3+/Mn4+ redox potential, high energy densities and high power densities. Lithium- ion batteries are much dependent on cathode material insertion-deinsertion processes. LiMn2O4, LiCoO2 are used as commercial positive cathode materials for various applications and high-voltage spinel cathode materials are also used for various energy-storage devices. Among all the positive materials used for lithium-ion battery applications, LiMn2O4 is an apt cathode material for rechargeable lithium-ion batteries, owing to its low cost, easy availability, environmentally benign nature, non-toxicity, and ease of synthesis while comparing with other layered oxides such as LiCoO2 and LiNiO2  -  . However, specific capacity of undoped LiMn2O4 decreases gradually upon repeated cycling carried out at elevated temperature   . It is well known that the capacity fading is caused due to several factors such as Jahn-Teller distortion, two-phase unstable reaction  , slow dissolution of manganese into the electrolyte  , lattice instability  , and particle size distribution   . In order to overcome the problem of Jahn-Teller distortion for obtaining the high capacity retention, several researchers have investigated earlier lithium rich spinels with various divalent, trivalent and tetravalent-doped ions such as Cr, Fe, Zn, Cu, Ga, Co, Al, Ni and Ti  . Ohzuku et al.  and Lee et al.  showed that partial doping of cations is effective in suppressing the capacity fade upon cycling. Moreover, the capacity fade of LiMn2O4 often happens much in 3 V region which can be completely suppressed by doping selenium with LiMn2O4  . Low temperature synthesis methods viz., sol-gel   , chemical precipitation  , hydrothermal and pechini process  have been used to obtain cathode materials with expected physical and electrochemical properties to use in lithium-ion batteries. Spinel LiMn2O4 and Zn, Co, Ni and substituted LiMn2O4 synthesized via facile sol-gel method to improve the electrochemical and structural properties of LiMn2O4 spinel based on electrode materials for Li-ion batteries  . The present work highlights that efforts have been taken to synthesize physical and electrochemical characterization of LiMn2O4 and LiCuxCryMn2−x−yO4 (x = 0.50; y = 0.05 - 0.50) via sol-gel method using Myristic acid as new chelating agent for the exploration of this facile synthesis method with good electrochemical performance.
LiMn2O4 and LiCuxCryMn2−x−yO4 (x = 0.50; y = 0.05 - 0.50) powders have been synthesized via sol-gel method using Myristic acid as chelating agent. Figure 1 depicts the flow chart of the synthesis procedure. Stoichiometric amounts of acetates of lithium, manganese, and the dopant salts such as acetate of copper and chromium were dissolved independently in triple-distilled water. The chelating agent (Myristic acid) is dissolved slowly in heating ethyl alcohol gently. After dissolving all the salts gently, it will be mixed with chelating agent. A small amount of precursor has been taken for TG/DTA analysis to under the thermal behaviour and the rest of the sample is calcined at 850˚C for 6 h. For thermal analysis, the precursors are heated in air atmosphere at 10˚C/min to 850˚C. All the calcined samples are subjected for physical characterization using thermo gravimetric/differential thermal analysis (TG/DTA-Seiko Exstart 6000, Japan), X-ray diffraction (XRD-Bruker D2 Phaser desktop, Cu source, AXS, Karisruhe, Germany), Fourier-transform infrared spectroscopy (FTIR-Tensor 27, Bruker (Germany)), field emission scanning electron microscopy (FESEM-JEOL, Model-JSM-7000F (Japan)), transmission electron microscopy (TEM-JEOL, JEM-ARM 200F, Japan), energy dispersive X-ray analysis (EDAX-JEOL, Model-JSM-7000F, Japan) and galvanostatic charge-discharge cycling studies (WonATech- Model, WBCS 3000, South Korea), electrochemical impedance spectroscopy (EIS-Potentiostat/Galva-nostat, Model-273A, USA) and differential capacity curves (dQ/dE).
Coin cells of 2032 configuration have been assembled in an argon filled glove box (MBraun, Germany) using lithium foil as anode, Celgard 2400 as separator, 1 M solution of LiPF6 and 0.3 M LiBF4 in 3:7 (v/v) volume mixture of ethylene carbonate (EC) and diethyl carbonate (DEC) as electrolyte and the synthesized material has been as cathode. The cathode is prepared by slurry coating procedure from a mix comprising synthesized compound, Super P carbon black as conducting material and poly (vinylidiene fluoride) 5% PVdF binder in n-me- thyl-2-pyrrolidone (NMP) solution mixed in the percentage ratio of 80:10:10 so as to form slurry. The slurry has been coated over aluminum foil and vacuum dried at 110˚C for 2 h. Electrode blanks of 18 mm diameter
Figure 1. Flow chart for synthesis of LiCuxCryMn2−x−yO4 by sol-gel method using Myristic acid as chelating agent.
are punched out and used as cathode in the coin cell. The fabricated 2032 coin cells have been cycled at a constant current of C/10 rate between 3.0 to 4.5 V using an in-house battery cycling unit.
3. Results and Discussion
3.1. Thermal Studies
The TG/DTA curve of LiMn2O4 precursor is depicted in Figure 2(a). The TGA curve clearly illustrates the three weight loss regions. Ab initio, the low weight loss of 5% is seen up to 100˚C owing to removal of water. Further, another two regions are observed between 100 V and 350˚C extending with maximum weight loss of 45% may be assigned to the decomposition of chelating agent (Myristic acid) and acetate precursors. It is well known that
(a)(b) (c)(d) (e)
Figure 2. Thermo gravimetric and differential thermal analysis (TG/DTA) of LiMn2O4 and LiCuxCryMn2−x−yO4 spinel precursor. (a) Undoped; (b) Cu: 0.50; (c) Cu: 0.50; Cr: 0.05; (d) Cu: 0.50; Cr: 0.10 and (e) Cu: 0.50; Cr: 0.50.
in the case of DTA curve, it depicts lucidly a well defined exothermic peak at 327.4˚C indicating the formation of the spinel compound which is mirrored in the XRD pattern corresponding to high degree of crystallinity for the sample calcined at 850˚C. Furthermore, all the peak reflections of LiMn2O4 have been shown unclouded when the precursor is calcined at 250˚C and 450˚C. Also, the TGA curve shows that the temperature above 350˚C stops totally thermal events without any further thermal reaction after the formation of spinel compounds.
TG/DTA curves of spinel LiCuxCryMn2−x−yO4 (x = 0.50; y = 0.05 - 0.50) precursors have been synthesized via sol-gel method using Myristic acid as chelating agent are presented in Figures 2(b)-(e). TGA curve of copper doped spinel precursor shows three characteristic weight loss zones (Figure 2(b)). Initially, the low weight loss of 5% up to 100˚C is an account of removal of moisture. The second and third weight loss zone of 40% up to 300˚C is attributed to the decomposition of acetate precursor of copper. DTA curve shows a well defined an exothermic peak at 286.3˚C suggesting the formation of spinel product. DTA curve corroborates no further thermal reactions are taking place beyond 300˚C.
Moreover, in the case of dual doped spinel (LiCuxCryMn2−x−yO4) precursor, LiCu0.5Cr0.05Mn1.45O4 shows the same three weight loss zones with weight loss of 5% compared to other all dual dopant concentration of the spinel precursors. This low amount in weight loss has been reflected in charge-discharge study to deliver the high discharge capacity of the spinel. Among all dual doped spinel precursors, equal doping ratio or Cu-Cr high doping of LiCu0.5Cr0.5MnO4, shows a well defined exothermic peak (see Figure 2(e)) at 307.4˚C with higher formation temperature owing to higher specific heat of Cr than Cu and Mn (Cr = 0.46 kJ/kg K; Cu = 0.39 kJ/kg K; Mn = 0.48 kJ/kg K).
Hence, in all cases, the weight loss zones complete up to 100˚C may be attributed to the removal of moisture and resulting in the formation of spinel at around 307.4˚C - 327.4˚C. This temperature difference (20˚C) is depicted in the formation temperature of all doped precursors which leads the reaction to begin much earlier and involving higher heat energy (307.4˚C) as indicated in the DTA curve (Figure 2(e)). All precursors show lucidly thermally inactive regions beyond 300˚C indicating the closure of thermal events.
3.2. X-Ray Diffraction Studies
Figures 3(a)-(e) depicts the XRD patterns of LiMn2O4 samples calcined at different temperatures: (a) as synthesized; (b) 250˚C; (c) 450˚C; (d) 650˚C; (e) 850˚C. It is evident that as synthesized and the samples calcined at 250˚C exhibits broad and indistinct reflections indicating the nebulous nature of the compounds. Moreover, the sample calcined at 450˚C and 650˚C exhibits an additional impurity peaks may be attributed to the formation of α-Mn2O3 and LiMn2O3. Similarly, the high intensity peaks such as (111), (311), (222), (400), (331), (551), (440) and (351) have been obtained for the samples calcined at 850˚C for 6 h confirms the high degree of crystallinity with better phase purity. These planes are in good agreement with the results obtained for the spinel compound synthesized via sol-gel method or solid-state method  -  . It is well known that the spinel compound has an Fd3m space group wherein lithium occupies at 8a tetrahedral sites, manganese and dopant ions occupy the
Figure 3. XRD patterns of sol-gel-synthesized undoped LiMn2O4 samples calcined at different temperatures viz., (a) As synthesized; (b) 250˚C; (c) 450˚C; (d) 650˚C; and (e) 850˚C.
16d sites and oxygen at 32e sites. All the XRD peak reflections for synthesized spinel match perfectly with Joint committee on Powder Diffraction Standard (JCPDS card No. 35-782).
The XRD patterns of LiCuxCryMn2−x−yO4 are shown in Figures 4(a)-(d) with different stoichiometric amounts of divalent and trivalent metal cations viz., Cu and Cr (Cu-0.5; Cr-0.05-0.5), synthesized via sol-gel method calcined at 850˚C. LiMn2O4 and LiCuxCryMn2−x−yO4 depict the formation of phase pure crystalline spinels in substantiating the first order reaction. Moreover, it is clear that all the peaks corresponding to (111), (311), (222), (400), (331), (551), (440) and (351) which are in good accordance with other researchers  . All the XRD peak reflections for this pristine spinel perfectly match with Joint committee on Powder Diffraction Standard (JCPDS card No. 35-782).
3.3. FTIR Spectroscopy
The FT-IR spectra of sol-gel synthesized LiMn2O4 powders calcined at different temperatures viz., 250˚C, 450˚C, 650˚C and 850˚C are shown in Figures 5(a)-(d). It is clear that the broad peaks appear for the sample calcined at low temperature (250˚C, 450˚C and 650˚C) for the sake of the water molecules (O-H bond). FT-IR studies on LiAlxMn2−xO4 have been investigated through solid-state combustion synthesis method in which all the frequencies showing striking similarity with our present investigation  . It is ratified that in the case of black coloured spinel compound obtained via sol-gel method exhibiting the two IR spectral bands at wavelengths between 513 - 516 cm−1 and 607 - 610 cm−1 for the sample calcined at 850˚C could be assignable to the Li-O bending vibration mode and Li-Mn-O stretching vibration band. The sample calcined at low temperature (250˚C) depicting the low wave number while comparing to undoped spinel calcined at 850˚C. The wave number corresponding to the LiMn2O4 and LiCuxCryMn2−x−yO4 spinels are given in Table 1 and Table 2. Also, the sample calcined at 850˚C depicts the IR band at around 610 cm−1 which is slightly shifted towards the lower wave number.
Figure 4. XRD patterns of LiCuxCryMn2−x−yO4 powders calcined at 850˚C. (a) Cu: 0.50; (b) Cu: 0.50; Cr: 0.05; (c) Cu: 0.50; Cr: 0.10 and (d) Cu: 0.50; Cr: 0.50.
Table 1. FTIR frequencies for the peaks observed for LiMn2O4.
(a) (b)(c) (d)
Figure 5. FT-IR spectra of spinel LiMn2O4 (a)-(d) particles calcined at different temperatures viz., 250˚C, 450˚C, 650˚C and 850˚C.
Table 2. FTIR frequencies for the peaks observed for LiCuxCryMn2−x−yO4.
Figures 6(a)-(d) depict the FT-IR spectra of LiCuxCryMn2-x-yO4powders with varying amounts of Cu and Cr. It is clearly shown that the di and trivalent doped spinels exhibiting the two IR spectral bands between 523 - 525 cm−1 and 618 - 625 cm−1 may be attributed to the Li−O bending vibration at lower wave number and Li-Cu- Cr-Mn-O stretching vibration at higher wave number respectively. The FT-IR on LiCrxNiyMn2−x−yO4 reveals the synthesis process via sol-gel method  . These results are is in good agreement with our FTIR data. Similarly, LiCrxCuyMnx−yO4 has been synthesized via wet chemistry method and reveals the spectra of stretching and bending vibration  . Such reports show the striking similarity in our present investigation. Nevertheless, it is corroborated that there are no any significant impurity peak reflections in all samples were calcined at 850˚C. High equal amount of dual doping spinel (LiCu0.5Cr0.5MnO4) exhibits the spectral bands 525 cm−1 and 625 cm−1 which is similar to wave number of 0.5-Cu doped spinel.
(a) (b)(c) (d)
Figure 6. FT-IR spectra of LiCuxCryMn2−x−yO4 particles with varying Cu-Cr doping calcined at 850˚C. (a) x = 0.50; (b) x = 0.50; y = 0.05; (c) x = 0.50; y = 0.10; (d) x = 0.50; y = 0.50.
3.4. FESEM Analysis
Figures 7(a)-(e) depict the Field Emission Scanning Electron Microscope images of undoped and LiCuxCryMn2−x−yO4 spinel powders calcined at 850˚C. The undoped pristine spinel depicts cauliflower morphology with an average particle size of 50 nm (Figure 7(a)) with good agglomerated particles. Similarly in the case of Cu doped samples, it is lucidly seen pebbles morphology seems to be 2 µm. It is evident that 0.05-Cr (Figure 7(c)) doped spinel depicts spherical surface morphology with less particle size of 0.50 µm. The equal doping ratio of LiCu0.50Cr0.50Mn1.0O4 shows clearly that all the large particles are seen to be spherical grains haphazardly with an increased particle size of 2 µm as it possesses high amount of Cu-Cr doping which is reflected in electrochemical impedance spectroscopy (EIS) and cycling behavior of the doped spinel.
3.5. TEM Analysis
Figures 8(a)-(f) depict the TEM images of LiMn2O4 and LiCuxCryMn2−x−yO4 particles calcined at 850˚C. Figure 8(a) illustrates the selected area of diffraction pattern (LiMn2O4) suggesting the diffuse hollow with multiple fringes. The undoped spinel shows (Figure 8(b)) uniform spherical morphology with particle size of (100 nm). In the case of LiCu0.5Mn1.5O4, (Figure 8(c)) all the particles are seen like large spherical surface morphology. Moreover, LiCu0.5Cr0.05Mn1.45O4 (Figure 8(d)), depicts cloudy particles with good agglomerated particle size of 100 nm. This good agglomeration has been lucidly resulted to increase the high the electrochemical activity of the spinel among all dopants. Therefore, the high equal amount of dual doping spinel (LiCu0.5Cr0.5MnO4) (Figure 8(f)) depicts the typical morphology with large particles since chromium content is higher than that of other dopant which is unclouded reflected in charge-discharge studies leading to rapid capacity fading upon the cycling.
3.6. EDAX Analysis
Figures 9(a)-(e) depict the EDAX peaks of Cu, Cr, Mn and O in LiMn2O4 and LiCuxCryMn2−x−yO4 compounds.
(a) (b)(c) (d)(e)
Figure 7. FESEM images of spinel LiMn2O4 and LiCuxCryMn2−x−yO4 particles calcined at 850˚C. (a) Undoped; (b) Cu: 0.50; (c) Cu: 0.50; Cr: 0.05, (d) Cu: 0.50; Cr: 0.10 and (e) Cu: 0.50; Cr: 0.50.
All the EDAX peaks corroborate lucidly the actual compositions of the undoped and doped spinels to be pure without any additional impurities. Table 3 depicts the EDAX compositions of various elements in LiMn2O4 and LiCuxCryMn2−x−yO4.
3.7. Galvanostatic Charge-Discharge Studies
Figures 10(a)-(e) show the first cycle charge-discharge behavior of LiMn2O4 and LiCuxCryMn2−x−yO4 with different stoichiometric concentration of Cu and Cr. The charge-discharge behavior of undoped and LiCuxCryMn2−x−yO4 shows pellucid the de-intercalation/intercalation of lithium ions at the 8a tetrahedral sites from cubic spinel structure. At the beginning, LiMn2O4 has been synthesized via sol-gel method and calcined at 850˚C delivers the high discharge capacity of 130 mA∙h/g against the charging capacity of 138 mA∙h/g with
(a) (b) (c)(d) (e) (f)
Figure 8. TEM images of LiMn2O4 and LiCuxCryMn2−x−yO4 particles calcined at 850˚C. (a) (b) Undoped; (c) Cu: 0.50; (d) Cu: 0.50; Cr: 0.05; (e) Cu: 0.50; Cr: 0.10 and (f) Cu: 0.50; Cr: 0.50.
Table 3. EDAX compositions of various elements in LiMn2O4 and LiCuxCryMn2−x−yO4.
(a)(b) (c)(d) (e)
Figure 9. EDAX images of LiMn2O4 and LiCuxCryMn2−x−yO4 powders calcined at 850˚C. (a) Undoped; (b) x = 0.50; (c) x = 0.50; y = 0.05; (d) x = 0.50; y = 0.10; (e) x = 0.50; y = 0.50.
cathodic efficiency of 88% corresponds to 94% columbic efficiency during the first cycle. It is evident to note that the undoped spinel exhibits superior performance to other doped spinels (LiCuxCryMn2−x−yO4). Moreover, undoped spinel has been derived via sol-gel method  and delivers the maximum discharge capacity of 125
(a)(b) (c)(d) (e)
Figure 10. First cycle charge-discharge behaviour of LiMn2O4 and LiCuxCryMn2-x-yO4. (a) Undoped; (b) x = 0.50; (c) x = 0.50; y = 0.05; (d) x = 0.50; y = 0.10; (e) x = 0.50; y = 0.50.
mA∙h/g during the first cycle in which these results showing inferior performance to our present investigation. Similarly, another pristine LiMn2O4 has been synthesized through sol-gel method offers an initial reversible capacity of 128 mA∙h/g  wherein it seems to be lower than our result. Also, copper doped spinel has been attempted to synthesize via sol-gel method  and delivers the maximum discharge of capacity of 128 mA∙h/g during the first cycle. These experiment results are slightly lower than to our present report. Nevertheless, during sol-gel synthesis of spinel compounds, Myristic acid has been used as chelating agent to be favorable and to act as catalyst to fasten the reaction causing the multi-ligand chain between Mn-O and COO− resulting in the formation of pristine LiMn2O4 and LiCuxCryMn2−x−yO4 for augmenting the high electrochemical stability of the spinel compound.
Figures 10(b)-(e) show the first cycle charge-discharge behavior of LiCuxCryMn2−x−yO4 heated at 850˚C. Copper and copper-chromium doped spinels exhibit the discharge capacity of 122, 124, 120 and 120 mA∙h/g corresponds to 77%, 78%, 77% and 76% columbic efficiency during the first cycle (Figures 10(b)-(e)). It is quite interesting to note that 0.5-Cu doped spinel has been experimented to synthesize through sol-gel route  and yields the maximum discharge capacity of 70 mA∙h/ g in the first cycle. It shows the poor performance rather than our results. Similarly, 0.05-Cr3+ modified LiMn2O4 spinel intercalation cathodes through oxalic acid assisted sol-gel method for lithium rechargeable batteries has been reported  and exhibits an initial discharge capacity of 80 mA∙h/g wherein these results are lower than that of our present performance. It is well known that the sample with low Cr-doping (LiCu0.5Cr0.05Mn1.45O4) exhibits the higher discharge capacity (124 mA∙h/g) than the other all slightly high doping compositions. The increase in capacity may be due to the same oxidation state of chromium ions have smaller ionic radii than manganese ions, Cr3+ (0.615 Å), Mn3+ (0.68 Å), Cr4+ (0.58 Å), Mn4+ (0.60 Å). Also the stronger Cr-O bonds in the delithiated state (compare the binding energy of 1142 kJ∙mol−1 for CrO2 with 946 kJ mol−1 for α-MnO2) may also be considered to contribute to stabilization energy of Cr3+ in the octahedral sites. Therefore, the equal doping ratio of spinel (LiCu0.5Cr0.5Mn1.0O4) delivers slightly lower performance (discharge capacity of 120 mA∙h/g) than undoped and other dopant compositions may be owing to high order of cationic mixing of dopants, poor electronic conductivity, and higher electrochemical impedance has been developed inside the cell which is reflected in electrochemical impedance spectra (ECI).
The cycling behavior of undoped and doped spinels calcined at 850˚C over investigated 10 cycles with the corresponding columbic efficiencies (CE) is shown in Figures 11(a)-(e). The undoped spinel exhibits the maximum discharge capacity of 130 mA∙h/g in the first cycle. However, it besets from capacity fading drastically upon cycling up to 10 cycles and also it experiences a capacity fade of 1.4 mA∙h/g cycle over the investigated 10 cycles with capacity retention of 89%. In the 5th and 10th cycle, the undoped spinel delivers a discharge capacity of 125, 116 mA∙h/g or both same columbic efficiency of 99%. Similarly, the cells with copper and copper- chromim doped spinels deliver the discharge capacities of 123, 124, 120 and 120 mA∙h/g during the first cycle and experiencing a capacity fade of 1.1, 1.1, 1.2 and 1.4 mA∙h/g cycle with capacity retention of 92, 90, 91 and 88% for the Cu and Cu:Cr contents corresponding to 0.5, 0.5; 0.05, 0.5; 0.10 and 0.50; 0.50, respectively over the investigated 10 cycles. In the 5th and 10th cycle of doped spinels, the discharge capacities have been exhibited 118, 112; 115, 111; 112, 109; 114, 106 mA∙h/g; with capacity fades of 0.1, 0.2; 0.1, 0.1; 0.1, 0.1; 0.1, 0.2 mA∙h/g cycle corresponding to columbic efficiency of 99v, 98%; 99%, 99%; 99%, 99%; 99%, 98% for the Cu and Cu:Cr contents with regard to 0.5, 0.5:0.05, 0.5:0.10 and 0.50:0.50, respectively. Inter alia all dual doped compositions, the low Cr-doped spinels (LiCu0.5Cr0.05Mn1.45O4) offers better stable capacity retention up to 10 cycles. Several researchers have already reported  that 0.2-Cr doped LiMn2O4 has been synthesized by sol-gel method as cathodes in high-voltage lithium batteries and delivered the maximum discharge capacity of 78 mA∙h/gin the 10th cycle. These results are not encouraging and inferior to our present investigation (109 mA∙h/g). Also, 0.01-Cr doped spinel has been synthesized towards intercalation cathodes through oxalic acid assisted sol-gel method for lithium rechargeable batteries and delivers the inferior results of discharge capacity of 115 mA∙h/g in the 10th cycle  when compare to our present investigation. Moreover, 0.01-Cr doped pristine spinel has also been prepared using Adipic acid as chelating agent via sol−gel route and exhibits the maximum discharge capacity of 117 mA∙h/g in the 10th cycle  wherein these results are found to be somewhat lower in our earlier experiment rather than this work. In other words, 0.5-Cu doped spinel has been reported via sol-gel method  and delivers the maximum discharge capacity of 60 mA∙h/g in the 10th cycle which is two times lower than our results. Moreover, 0.1-Al doped spinel has been experimented via traditional sol-gel method  exhibits 100 mA∙h/g in the 10th cycle sans stable discharge capacity wherein these results are lower
(a) (b)(c) (d)(e)
Figure 11. Cycling behaviour of LiMn2O4 and LiCuxCryMn2−x−yO4. (a) Undoped; (b) x = 0.50; (c) x = 0.50; y = 0.05; (d) x = 0.50; y = 0.10; (e) x = 0.50; y = 0.50.
than our reported investigation (106 mA∙h/g) with good capacity retention. It is quite interesting to know the discharge capacities of pristine spinels synthesized by different methods suggesting that 0.1-Cr doped spinel synthesizes via green chemistry method/sol-gel method and delivers the specific discharge capacity of 95 mA∙h/g in the 10th cycle  . These results are not superior to our present work. Hence, it is concluded from the charge-discharge and cycling studies that among all dual doped spinels, the low amount of Cr-doped spinel (LiCu0.5Cr0.05Mn1.45O4) is an apt candidate to enhance the electrochemical stability of the spinel owing to higher octahedral stabilization energy of Cr3+ (1142 kJ∙mole−1) as compared to that of manganese (946 kJ∙mole−1) and effect of chromium will be more pronounced in reducing the capacity fade leads to decrease in cell volume to augment the stability of the structure during the deintercalation/intercalation process. Thus, low chromium doped spinel (LiCu0.5Cr0.05Mn1.45O4) stabilizes the Mn3+ structure for increasing good cycleability with better capacity retention. Also the stronger Cr-O bonds in the delithiated state (compare the binding energy of 1142 kJ∙mol−1 for CrO2 with 946 kJ∙mol−1 for α-MnO2) may also be considered to contribute to stabilization energy of Cr3+ in the octahedral sites.
3.8. Electrochemical Impedance Spectroscopy
Figures 12(a)-(e) show the Nyquist plot of electrochemical impedance spectra of (EIS) coin cells after 10 cycles. The Nyquist plots depict the single semicircle in the high frequency region and straight sloping line in the low frequency region. Such Nyquist plots are the relaxation frequency due to the interfacial polarization. All the Nyquist spectra depict the high-frequency semicircle with a maximum at a frequency of few kilo-ohms. This may be due to highly charged state and the typical behavior of the diffusion kinetic process inside the electrode materials. Also, having the high-frequency semicircle to the charge of the surface area of the spinel oxide positive electrodes. This real impedance belongs to both interparticle electronic contact and ionic diffusion via the passivation layer. It is clearly seen that among all the five compounds, LiCu0.5Cr0.05Mn1.45O4 shows low resistance of 190 Ohms while undoped spinel depicting the real impedance of 350 Ohms. Similarly, in the case of LiCu0.5Mn1.5O4, LiCu0.5Cr0.1Mn1.4O4 provides higher resistance of 500 and 275 Ohms than that of LiCu0.5Cr0.05Mn1.45O4. Moreover, an equal amount of high doping pristine spinel (LiCu0.5Cr0.5Mn1.0O4) exhibits the high electrochemical polarization of 2000 Ohms which is seemed to be higher impedance relatively to all the dopants. During charging, Li/LiCu0.5Cr0.05Mn1.45O4 couple delivers very lower electrochemical polarization of 350 Ohms than that of undoped and all doped spinels for the sake of good kinetic reactions and good thermodynamics. Hence, among all dual doped spinels, the stable specific discharge capacity has been offered by the low Cr-doped spinel (LiCu0.5Cr0.05Mn1.45O4) may be due to the low order of electrochemical polarization/charge transfer resistance (Rct) and better electronic conductivity with mitigating of Mn3+ from Mn4+ for high diffusion of lithium ions during intercalation and de-intercalation process. LiCu0.5Cr0.5Mn1.0O4 samples show higher electrochemical impedance as a consequence of increasing the charge transfer resistance and charge transfer of lithium ions on the surface of the electrodes which leads to capacity fading upon cycling (10 cycles). In view of the above, the impedance behaviour of 0.05-Cr doped spinel shows an excellent electrochemical performance.
3.9. dQ/dE vs. Potential Curves
The differential capacity curve of the parent LiMn2O4calcined at 850˚C is shown in Figure 13(a). This curve depicts lucidly two redox peaks respectively. The oxidative peaks correspond to the extraction of lithium ions and reductive peaks are insertion of lithium ions. The first two predominant anodic peaks are seen at about 4.01 and 4.15 V corresponds to Mn3+/Mn4+ couples with an average oxidation state of 3.5 while another two well defined cathodic peaks appearing at around 3.98 and 4.1 V owing to the reductive behavior of the spinel. Moreover, these two anodic peaks of undoped spinel depict the high anodic peak currents at around 0.0018 and 0.00145 mA∙hg−1∙mV−1 with the cathodic peak currents of −0.00195 and −0.00194 mA∙hg−1∙mV−1. Hence, it is vindicated that the anodic process is more facile than cathodic one.
Figure 13(b) depicts the differential capacity cure of LiCu0.5Cr0.05Mn1.45O4 calcined at 850˚C. It is well known that two well defined anodic peaks are seen between 4.02 and 4.15 V attributed to the de-intercalation process with Mn3+/Mn4+, Cu2+/Cu3+ and Cr3+/Cr4+ couples respectively. Similarly, two anodic peaks of LiCu0.5Cr0.05Mn1.45O4 exhibit the high anodic peak currents at around 0.0065 and 0.002 mA∙hg−1∙mV−1 with the cathodic peak currents of −0.0045 and −0.0021 mA∙hg−1∙mV−1 at about 4.11 and 4 V. It is evident that the differential capacity curve of LiCu0.5Cr0.05Mn1.45O4 exhibits the better performance in both peak currents when
(a)(b) (c)(d) (e)
Figure 12. Electrochemical Impedance spectra (EIS) of the cells (LiMn2O4 and LiCuxCryMn2−x−yO4) after 10 cycles. (a) Undoped; (b) x = 0.50; (c) x = 0.50; y = 0.05; (d) x = 0.50; y = 0.10; (e) x = 0.50; y = 0.50.
Figure 13. Differential capacity curves of (a) LiMn2O4; (b) LiCu0.5Cr0.05Mn1.45O4.
compared with undoped spinel. Hence, it is concluded that the low amount of chromium doped spinel (LiCu0.5Cr0.05Mn1.45O4) increases the electrochemical stability of the compound which has been reverberated in cycleability studies.
Pristine spinel LiMn2O4 and LiCuxCryMn2−x−yO4 (x, y = Cu, Cr) (x = 0.50; y = 0.05 - 0.50) powders have been synthesized via sol-gel method for the first time using Myristic acid as the chelating agent to obtain micron sized particles for use as cathode materials in lithium rechargeable batteries. XRD patterns of LiMn2O4 and LiCuxCryMn2−x−yO4 heated at 850˚C corroborate the high degree of crystallinity with better phase purity of synthesized materials via sol-gel synthesis. FESEM image of undoped pristine spinel lucidly reveals cauliflower morphology with good agglomerated particle size of 50 nm while 0.5-Cu doped samples depict the pebbles morphology. TEM images of the spinel LiMn2O4 and LiCu0.5Cr0.05Mn1.45O4 substantiate that all the synthesized particles via sol-gel method are nano-sized (100 nm) with spherical surface and cloudy particles morphology. EDAX peaks confirm their actual composition of Cu, Cr, Mn and O in LiMn2O4 and LiCuxCryMn2−x−yO4. Electrochemical impedance spectroscopy (EIS) measurements of spinel LiMn2O4 and LiCu0.5Cr0.05Mn1.45O4 exhibit the high and low order of polarization of (350 and 190 Ohms) after 10 cycles. Among all dopant concentrations attempted, LiCu0.5Cr0.05Mn1.45O4 sample exhibits the best performance (1st cycle discharge capacity: 124 mA∙h/g) with low capacity fade of 1.1 mA∙h/g cycle and capacity retention of 90%.
One of the authors Dr. R. Thirunakaran is thankful for the support given under the “Brain Pool Program of the Korean Federation of Science and Technology Societies” (KOFST), Republic of South Korea and also grateful to Prof. Won-Sub Yoon, for his help and valuable guidance. Further, Dr. R. Thirunakaran, thanks Dr. Vijayamohanan K. Pillai, Director, CSIR-CECRI for granting leave to avail the above fellowship. Also, this work was supported by the Human Resources Development Program (No. 20124010203270) of the Korea Institute of Energy Technology Evaluation and Planning (KETEP) grant funded by the Korea government Ministry of Trade, Industry and Energy. Many thanks are to the students of Professor for their co-operative helps and Cooperative Center for Research Facilities (CCRF) for completion of all physical characterization studies during my fellowship at Sungkyunkwan University (SKKU), South Korea.