Abstract
Au with Pd nanoparticles were synthesized and coated onto the spinel LiMn2O4 via a coprecipitation calcination method with the objective to improve the microstructure, conductivity, and electrochemical activities of pristine LiMn2O4. The novel composite cathode had high phase purity, well crystallized particles, and more regular morphological structures with narrow size distributions. At enlarged cycling potential ranges the sample delivered 90 mAh g−1 discharge capacity compared to LiMn2O4 (45 mAh g−1). It was concluded that even a small amount of the Pd and Au enhanced both the lithium diffusivity and electrochemical conductivity of the host sample due to the beneficial properties of their synergy.
1. Introduction
Lithium-ion batteries are becoming incredibly popular in modern electronic devices. Compared with traditional battery technology, lithium-ion batteries charge faster, with an operating voltage of ~3.7 V, last longer, and have a higher power density in a lighter package [1]. For the application of LIBs, the main concerns of electrochemical performance are cost reduction, cycle life, power density, energy density, and especially safety [2]. It has thus become particularly important to develop cathode materials with reversible and fast charge-transfer reactions in addition to high capacities. The most investigated and commercially used cathode materials include layered LiCoO2 and LiNiO2 [3], LiNiVO4 and LiCoVO4 [4], and LiMnO4 [5]. Among these the spinel LiMn2O4 provides several advantages for Li-ion batteries because of their high voltage (4 V versus Li/Li+), good cyclability, lower cost of production, lower toxicity, and safety [6]. Depending on the synthetic procedure, nanosized LiMn2O4 with different desirable physical and chemical properties can be produced [7, 8]. Solid-state reaction is the most conventional method to prepare LiMn2O4, however with the disadvantage of large particle size, wide-ranging particle size distribution, and irregular morphology [9]. To overcome these disadvantages, a sustainable effort of many researchers has gone into the development of soft chemistry methods [10]. Hence, the coprecipitation technique is considered the simplest and most efficient chemical pathway to yield nanocrystalline LiMn2O4. However, LiMn2O4 still suffers from structural reconstruction and Mn dissolution in the electrolyte upon cycling, which results in high capacity fading [11]. Therefore, most research attempts to improve the electrochemical performance of the spinel have been directed toward the synthesis of cation-doped LiMn2O4 and surface passivation treatment [12, 13]. The elements commonly used for modification include nonmetals, rare-earth metals, and actinide dopants. These modification methods can weaken the Jahn-Teller effect, decrease the dissolution of active material, minimize surface overpotential, and stabilize the structure which concomitantly enhances the electrochemical performance of the cathode [14]. In this study, palladium is of interest from both fundamental and technological viewpoints because of quantum size effect, which is derived from the reduction of free electrons [15]. Likewise, gold and palladium bimetallic nanoparticles [16] have recently emerged as viable catalysts due to their unique properties. These properties include large binding energy and the potential to reduce Mn3+ content, increase the average valence of Mn, and stabilize the cubic structure of LiMn2O4 compared to their bulk counterparts. In this communication, the use and beneficial synergistic effects of Pd-Au alloy as a novel LiMn2O4 surface coating material are reported. The correlation of Li[PdAu]xMn2−xO4 on the electrochemical performance, crystal structure stability, and morphology is investigated and discussed.
2. Experimental Section
2.1. Synthetic Methodology of Cathode Composite
(a) Synthesis procedure: the LiMn2O4 cathode nanomaterial with spherical nanostructures was successfully synthesized by coprecipitation method from the reaction of lithium hydroxide and manganese acetate. A stoichiometric amount of LiOH and Mn(CH3COO)2 with the cationic ratio of Li/Mn = 1 : 2 were dissolved in deionized water by stirring gently. The solution was evaporated at 100°C for 10 h to obtain the precursor powder. The precursor was calcined at 600°C for 10 h to form the semicrystallite LiMn2O4 powder. The product was subjected to acid treatment in 2 M H2SO4 for 2 h to increase the degree of oxidation and remove remnant Mn2O3 and Mn3O4. (b) Coating strategy:the Pd-based bimetallic nanoparticles were prepared via an emulsion-assisted synthetic strategy. For surface modification, the Pd-Au nanoparticles and crystalline LiMn2O4 were added to deionized water and heated at 100°C under stirring until the solvent fully evaporated [17]. The well mixed powder was then calcinated at 500°C for 10 h in air. The reaction forms a solid composite of Li(M)xMn2−xO4 () which contributes to structural stability [18].
2.2. Preparation of LiMxMn2−xO4 Cathode Coin Cells
The cathode was prepared by mixing 80 wt% of the pure and Li[PdAu]0.02Mn1.98O4 active composite material, respectively, with 15 wt% acetylene black (current collector) and 5 wt% polyvinylidene fluoride binder (PVDF, dissolved in N-methyl-2-pyrrolidone, NMP) to form a slurry. The typical mass loading of active material was ~2 mg/cm2. The slurry was then cast on the aluminium foil and cathode disks with a diameter of 12 mm were punched out and dried for 24 hours at 120°C. Test cells (LR2032, 20 d × 3.2 mm) were assembled and sealed in an Argon-filled glove box. The electrolyte used was 1 M LiPF6 in ethylene carbonate (EC) : dimethyl carbonate (DMC) in a 1 : 1 volume ratio and Celgard 2400 polyethylene/polypropylene as the separator.
2.3. Characterization
The shape and size of the microstructure of Li[PdAu]0.02Mn1.98O4 cathode were observed using a Hitachi model X-650 scanning electron microanalyser and a Tecnai G2 F20 X-Twin MAT 200 kV transmission electron microscope. The samples were characterized by X-ray powder diffraction (XRD), which was recorder on a BRUKER AXS (Germany), D8 Advance diffractometer, using Cu-Kα radiation (λKα1 = 1.5406 Å). The diffraction patterns were taken at room temperature in the range of 5 < 2θ < 90° using step scans. All electrochemical measurements used for characterization were carried out using a coin cell. Electrochemical impedance measurements were recorded with a Zahner IM6ex (Germany) workstation, at a perturbation amplitude of 10 mV within a frequency range of 100 kHz to 100 mHz. All potentials given in this paper refer to Li/Li+.
3. Results and Discussions
The SEM micrograph of the spinel LiMn2O4 is shown in Figure 1. The pure LiMn2O4 have apparent primary particles around 50 nm and have the characteristic spinel shape. Figure 1(a) shows the secondary particles of LiMn2O4 to be about 100 nm, which are glomeration congregated tightly by primary particles, indicating that the crystals of the spinel LiMn2O4 grow very well. Well dispersed PdAu nanoparticles across the LiMn2O4 surface are shown in Figure 1(b). The Li[PdAu]0.02Mn1.98O4 nanoparticles retained a well-developed octahedral structure with sharp edges after surface treatment with particle sizes in the range of 50–100 nm. The Li[PdAu]0.02Mn1.98O4, having an increased surface area, favors the penetration of electrolyte, decreasing the diffusion length of lithium ions. Detailed TEM and EDX analysis of the nanoparticles shown in Figure 2(a) appear to have almost uniform spherical shape, with a tight size distribution. Figure 2(b) is an enlargement of Li[PdAu]0.02Mn1.98O4 from which the particles appear highly crystalline. The nanocrystalline material admits electrolyte to allow rapid entry of lithium ions for quick battery charging and provide space to accommodate expansion and contraction during Li+ intercalation and deintercalation. The diffraction pattern of the coating layer revealed that it was a single crystal with cubic spinel structure as shown in Figure 2(b) (inset). Both the LiMn2O4 and Li[PdAu]0.02Mn1.98O4 cathodes adopt a typical spinel structure with Fd3m space group [19].
The structure of the Li[PdAu]0.02Mn1.98O4 nanoparticles was characterized by XRD as shown in Figure 3. The X-ray diffraction pattern of Li[PdAu]0.02Mn1.98O4 agrees with the pattern of pure LiMn2O4 and was identified as a single phase of cubic spinel with space group Fd3m. The sharp peaks and particle sizes of Li[PdAu]0.02Mn1.98O4 as probed by XRD correlate well with the TEM analysis. No addition peaks corresponding to that of the PdAu alloy were observed in the diffraction pattern of PdAu coated LiMn2O4. The formation of a Li[PdAu]0.02Mn1.98O4 composite structure was therefore signified by a shift (about 0.2°) in peak position to higher angles, with a slight lattice constant increase from 8.26006 Å to 8.26007 Å. This lattice expansion indicates that Pd-Au, having a larger ionic radius than Mn3+, partially diffused into the crystal structure of the spinel during heat-treatment [20, 21]. The small cell parameter changes in the lattice aid the stability of the structure and the improvement of the cycle life. Chan et al. observed a similar tendency, where the substitution of Mn3+ with Cu2+/Cu3+ and chromium caused an increase and decrease in lattice constant, respectively, due to the differences in ionic radii [22].
Figure 4 is the cyclic voltammograms for LiMn2O4 (a) and Li[PdAu]0.02Mn1.98O4 (b) cycled at 0.1 mV s−1. The peak separation difference suggests that the lithium intercalation and deintercalation process is much easier after modification. The two redox peak couples residing at 4.04/4.17 and 3.93/4.07 V are similar to features of voltammograms reported in [23]. The pair Pa1/Pc1 is attributed to the removal/addition of Li ions from/into half of the tetrahedral sites in which Li/Li interaction occurs. The Pa2/Pc2 pair is due to this process at the other tetrahedral sites in which lithium ions do not have any nearest neighbor Li-Li repulsive interactions [24]. These processes are accompanied by reversible Mn3+/Mn4+ redox reactions. For Li[PdAu]0.02Mn1.98O4, the shift in the anodic peak current to higher voltage is due to Pd-Au involvement in the prevention of Mn ion migration as a result of stronger bond strength compared to that of the Mn–O bond in the spinel structure [25]. The diffusion coefficient, , was 0.23 cm2/s. and 0.63 cm2/s. for samples (a) and (b), respectively. Faster lithium mobility in Li[PdAu]0.02Mn1.98O4 gives rise to better high rate performances. The increased peak currents noted for Li[PdAu]0.02Mn1.98O4 can be associated with lowered polarization and improved electrode kinetics in higher voltage region. The peak separation (0.2 V) of Li[PdAu]0.02Mn1.98O4 was less than that of LiMn2O4, suggesting that the lithium intercalation and deintercalation process is much more efficient due to the increased sample conductivity [26]. Likewise, the total impedance as shown in Figure 5 for Li[PdAu]0.02Mn1.98O4 was significantly lower than that of Li[Au]x Mn2−xO4 and LiMn2O4, confirming the integration of a conducting layer. A second semicircle starts disappearing at low frequency, indicating that the two-stage intercalation/deintercalation condition of lithium varies in Li[PdAu]0.02Mn1.98O4 [27]. The enhanced conductivity supports faster charge transportation at high current rates and is useful to prevent the pronounced pileup of Li+ ions and undesired Mn3+ ions on the surfaces during discharge [28, 29]. The lower value obtained for Li[PdAu]0.02Mn1.98O4 is suggestive of faster electron transfer kinetics. The capacitance, , value of samples (a), (b), and (c) increased steadily with values of 1.09, 2.45, and 3.05 µF. The latter is due to the presence of Pd-Au on the sample surface (in agreement with SEM observation) [30]. The exchange current i0 followed a similar trend, with values of 1.83 × 10−4, 2.75 × 10−4, and 8.32 × 10−4 A cm−2, which further confirmed the synergistic catalytic effect of Pd-Au with LiMn2O4.
To evaluate the cathode cycling behavior, conditions of up to 50 cycles were performed, using 0.1 C charge/discharge rates. Figure 6 shows the initial discharge capacity of Li[PdAu]0.02Mn1.98O4 at 160.0 mAh g−1 which drops to 149 mAh g−1 after the first 50 cycles. The Pd-Au content is therefore sufficient, as it does not form a barrier but instead promotes lithium-ion movement, which enhances the initial specific capacity [31]. The discharge curves for Li[PdAu]0.02Mn1.98O4 cell show a decrease of the effective capacity of the cell with increased cycling. This is called the capacity offset and the effect is common to most cell chemistries. The inset shows the charge/discharge curve of Li[PdAu]0.02Mn1.98O4 cycled at a current density of 14.8 mA g−1 (0.1 C rate) in the potential range 2.4/4.8 V. The specific capacity of LiMn2−xO4 coated with Pd-Aux (in smaller quantity) shows significant improvement in the cycling performance and activity when compared with other surface doping elements for LiMn2O4 [32, 33]. The two pseudoplateaus at around 3.9 and 4.2 V represent the (redox current peaks at the positively or negatively going voltammetry curves) electrochemical behaviour of spinel LiMn2O4 [34]. The respective discharge pattern correlates with the order of atom orientation in the structure which affects the charge/discharge cycling performance [35].
4. Conclusions
Novel transition metal alloy ( = Pd-Au) coated LiMn2O4 with improved high rate performances have been successfully designed and synthesized using a simple coating strategy. The present study is conceivably the first time that effects of Pd-Au transition metal coating on spinel-structured LiMn2O4 are being explored. The Li[PdAu]0.02Mn1.98O4 cathode exhibited improved rate capabilities and high rate cyclic performances compared to the pristine LiMn2O4. These improvements are due to the enhanced electronic conductivity and lithium diffusivity resulting from transition metal alloy coating. Although Pd and Au are deemed expensive for large scale industrial applications, they are uniquely durable, hence being economically viable. The [Pd-Au]-LiMn2O4 modification may be a suitable approach for producing lithium-ion battery cathodes with improved electrochemical characteristics.
Conflict of Interests
The authors declare that there is no conflict of interests regarding the publication of this paper.
Acknowledgment
This paper was financially supported by South Africa’s National Research Foundation (NRF).