Abstract
The present study demonstrates a novel approach by which titanium foils coated with LiMn2O4 nanocrystals can be processed into a high-surface-area electrode for rechargeable batteries. A detailed study has been performed to elucidate how surface morphology and redox reaction behaviors underlying these electrodes impact the cyclic and capacity behavior. These nanocrystals were synthesized by in situ sintering and exhibited a uniform size of ~55 nm. A direct deposition technique based on electrophoresis is employed to coat LiMn2O4 nanocrystals onto titanium substrates. From the analysis of the relevant electrochemical parameters, an intrinsic correlation between the cyclability and particle size has been deduced and explained in accordance with the Li intercalation/deintercalation process. Depending on the particle size incorporated on these electrodes, it is seen that in terms of capacitance fading, for nanoparticles cyclability is better than their micron-sized counterparts. It has been shown that electrodes based on such nanocrystalline thin film system can allow significant room for improvement in the cyclic performance at the electrode/electrolyte interface.
1. Introduction
Cubic spinel LiMn2O4 has been widely researched for its performance in secondary storage devices [1–4]. LiMn2O4 spinel structure consists of a cubic close-packed array of O2−, Mn3+, and possible Mn4+ that are located in the octahedral sites and Li+ in the tetrahedral site. The Mn3+ has an octahedral coordination to the oxygens, which forms three-dimensional MnO6 octahedra that act as host for Li+. The combination of these structural features makes spinel LiMn2O4 a stable compound in different electrolytes for storage applications. In a Lithium ion battery, during the electrochemical reactions, Li+ intercalates through the electrolyte, which involves three reaction stages, (i) Li+ diffusion within the electrode, (ii) Li+ transfer (charge transfer reaction) at the interface between the electrode and electrolyte, and (iii) finally the movement of Li+ in the electrolyte. The first stage is the rate-determining step, which governs the performance of the batteries that depends on both the phase and surface morphology of the electrode material. This means that by maintaining phase purity and effectively increasing their surface area through nano-structuring it is possible to enhance the kinetic properties of the electrode system by decreasing the diffusion length [5]. For fabricating high-performance LiMn2O4 electrodes, many high-temperature techniques have been reported, for example, pulsed laser deposition [6, 7], spray pyrolysis [8, 9], combustion method [10, 11], and so forth; however, not much has been reported on the electrophoretic deposition of LiMn2O4. One of the main reasons has been the requirement of high voltage in volatile electrolytes, and also it is difficult to deposit micron-sized particles due to their inherent size and mass. For instance, studies have shown [12] that at a high voltage of 400 V, submicronic LiMn2O4 can be electrophoretically deposited. However, at this high voltage especially with the acetone type electrolyte, the reaction can be hazardous and extremely volatile. The present study reports the first examples of direct deposition of LiMn2O4 nanocrystals as thin films onto Ti foils using a novel low-voltage (60 V) electrophoretic deposition technique. These electrodes exhibited a capacity of 147 mAh/g and capacity fade of 5% after 25 cycles.
2. Material and Methods
10 g of citric acid (C6H7O8 NICE chemicals, India) was completely dissolved in 10 mL of double distilled water. To this solution, 0.75 g of lithium carbonate (Li2CO3 NICE chemicals, India) and 9.8 g of manganese acetate (Mn(CH3COO)2 NICE chemicals, India) were added and completely dissolved by rigorously stirring for 30 min at room temperature. 0.2 mL of ethylene glycol (C2H8O2 NICE chemicals, India) was added to this solution. The pH of the solution was measured as 4. The pH of this solution was increased to 10 by dropwise addition of ammonia, resulting into a viscous brown-colored solution. This solution was kept for calcination by heating it at 140°C for 3 hr in air, after which the temperature was increased at a ramp rate of 5 C/min to 800°C for 3 hr. Furnace cooling was employed, after the calcination. The particle size and morphology of the resultant powder were analyzed using transmission electron microscopy (TEM,) and X-ray diffractometry (XRD, X’Pert PRO Analytical). The oxidation states of the synthesized powder were analyzed using X-ray photospectroscopy (XPS, Axis Ultra, Shimadzu).
The resultant LiMn2O4 nanopowders were used for electrophoretic deposition. For this an electrochemical setup was employed, comprising of titanium foil (1 cm × 1 cm × 0.2 mm) and platinum mesh used as an cathode and anode, respectively. Electrolyte consisted of solution of pure iso-propanol (20 mL). To this electrolyte, 12 mg of LiMn2O4 powders was dispersed uniformly under constant stirring. The deposition was carried out at 60 V for 3 hr at room temperature resulting in a thin uniform layer of LiMn2O4. Scanning electron microscopy (SEM, Model: JEOL JSM 6490 LA) was performed to analyze the surface of the electrodeposited LiMn2O4 layer. The particle size distribution was measured using Image J software from the TEM images. For electrophoretically deposited LiMn2O4, the thickness of the coating was measured using a surface profilometer (Veeco Dektak 150). Cyclic voltammetry (CV, electrochemical workstation: Newport Model) studies were done to evaluate the electrochemical performance of the electrophoretically deposited LiMn2O4 layer, which was kept as the working electrode in a three-electrode setup configuration. The reference and counterelectrode consisted of calomel and platinum. The electrolyte used for this purpose was 1 M lithium perchlorate in propylene carbonate.
3. Results and Discussion
Figure 1 shows the XRD pattern of the synthesized LiMn2O4 nanocrystals. Its diffraction peaks were perfectly indexed to a pure cubic spinel phase (JCPDS: 35–782). In addition, the broad diffraction peaks of the powder indicated that its particle size was small. Peak broadening in nanoparticles can originate from variations in lattice spacings, caused by lattice strain as the size decreases; the crystallite size was estimated as 40 ± 5 nm using the Scherrer equation [13].
TEM images displaying the morphology of LiMn2O4 are shown in Figure 2(a). It was found from high-resolution TEM (Figure 2(b)) that the interplanar spacings were about 0.6 nm showing a crystal orientation along the (111). This was confirmed by fast fourier transform (FFT) analysis (Figure 2(c)). The particle size analysis (Figure 2(d)) showed a skewed bimodal narrow distribution and was found to be in the range of 40–100 nm and primarily centered at ~55 nm.
(a)
(b)
(c)
(d)
The elemental composition and the valence states of the synthesized nanocrystals were done using XPS. Figure 3 gives the wide spectrum of LiMn2O4 crytals from 1200 to 200 eV. The presence of Li was distinctly detected in the powders at 54.5 eV. The Mn 2p peaks were deconvoluated (Figure 3 (inset)) at high resolution to reveal Mn 2p doublet for LiMn2O4. Similar patterns have been reported in the literature elsewhere [4, 14]. The Mn 2p core-level spectra showed a typical two-peak structure ( and ) due to the spin-orbit splitting. The XPS spectrum was calibrated to the C1s line, which was located at 285 eV. Using this reference, the peaks at 642.5 and 654.3 eV are attributed to Mn and Mn , respectively. Studies have shown [14, 15] that the Mn gives the XPS binding energy of Mn3+ and Mn4+ ions at 641.9 and 643.2 eV (in the present study this is shown as convoluted peaks in Figure 3 inset), respectively, which were in proximity to the values that were obtained in this study indicating the presence of mixed valence state of Mn ions in the synthesized nanocrystals.
The possible formation of stoichiometrically stable spinel LiMn2O4 (Fd3m is the space group and lattice parameter ) nanocrystals with the precursors used in the present study can be explained through the following reaction stages. Mn2+ undergoes hydrolysis resulting in MnOH+ [4] with the release of proton into the solution. MnOH+ are complexed by citric acid which chelates these ions at pH = 10. The possible esterification [16] occurs in the presence of ethylene glycol at 140°C, which results in gel formation. Further heating of this gel provides sites for the incorporation of Li+ into the Mn (II) complex to produce LiMn2O4 nanoparticles. The layout of possible chemical reactions for the above mechanism is shown in Figure 4. Figure 5(a) shows the surface morphology of the electrodeposited LiMn2O4 nanoparticle layer which exhibited a dense coating layer. Surface profilometry on this layer exhibited a highly roughened surface () and thickness of 5–8 μm (as shown in Figure 5(b)).
(a)
(b)
Figure 6(a) shows a typical cyclic voltammogram (scan rate: 1 mv/s) of the LiMn2O4 electrode. Four distinct peaks were identified on the CV pattern corresponding to different states of charge-discharge. These peaks were in accordance with the Li intercalation/deintercalation process and can be written as follows [17–21].(i)Reaction from (A) to (B):(ii)Reaction from (B) to (C):(iii)Reaction from (D) to (E):(iv)Reaction from (E) to (F):
(a)
(b)
From Figure 6(a) the total capacity was calculated as 147 mAh/g, which is comparable to conventional micron-sized and bulk LiMn2O4 systems in the literature [18–20].
The cycling ability of LiMn2O4 electrode was evaluated by the capacity fade, which was obtained from CV curves (scan rate: 50 mv/s) as shown in Figure 6(b). The result in Figure 6(b) reveals that the capacity fade of the electrode at the end of 25th cycle was ~5%. It was observed that the cycling ability of the electrophoretically deposited LiMn2O4 layer was stable, which can be attributed to the release of more Mn4+ cations that limit the Jahn-Teller distortion effect [22] encountered in octahedral complexes of the transition metals like those of LiMn2O4.
Further, the structural stability of the coating layer can also play an important role in determining the capacity of the electrode overlay. The formation of a thin and porous layer of LiMn2O4 can readily accommodate the microscopic stresses due to the any-dimensional changes within an electrode, for example, stresses arising during the phase transformation of cubic to tetragonal phase in LiMn2O4 during charge-discharge cycle.
Figure 7 shows the charge-discharge profile of LiMn2O4 at a discharging current density of 5 mA/g using 1 M lithium perchlorate in propylene carbonate. The capacity obtained from this curve was found to be ~80 mAh/g. It was found that when a micron-sized particulate structure of LiMn2O4 (~10 μm particle size) was deposited electrophoretically under similar conditions, the direct consequence was peeling and delamination of the coating layer during the charge-discharge cycles. Thus, for such an electrode system, the nanosize of LiMn2O4 structures was critical in determining the final performance of the electrode.
4. Conclusions
The phase-pure LiMn2O4 spinel nanocrystals of uniform size ~55 nm were synthesized at 800°C. XPS studies showed the presence of mixed valence state of Mn ions in the synthesized nanocrystals. The surface morphology of the electrodeposited LiMn2O4 nanoparticle layer exhibited a dense coating layer with a highly roughened surface () and thickness of 8–10 μm. CV studies showed four distinct peaks corresponding to different states of charge-discharge attributed to the Li intercalation/de-intercalation. The cyclability studies showed these electrodes to be stable, where the capacity fade of these electrodes at the end of the 25th cycle was found to be 5%. Charge the discharge profile of LiMn2O4 showed the capacity to be ~80 mAh/g. The present study opens an avenue for possible fabrication of thin film secondary storage batteries where LiMn2O4 nanocrystals can be deposited over scalable areas in a controlled fashion.
Acknowledgment
The Ministry of New Renewable Energy, Government of India is gratefully acknowledged for their financial support.