Introduction

The demand for sustainable and clean energy, such as hydrogen, wind, wave, and solar energies, is becoming more and more critical owing to the fossil fuel crisis [1,2,3,4,5,6]. Electrochemical energy storage and conversion devices such as batteries and supercapacitors play an extremely important role in the fields of portable electronic products, electric vehicles, and smart electrical grids [7,8,9,10,11]. Therefore, systems of electrochemical energy storage and conversion such as Li-ion batteries (LIBs), lithium oxygen batteries, lithium-sulfur batteries, fuel cells, and supercapacitors are highly desired to meet the growing requirement for a multitude of emerging applications [12,13,14,15,16,17,18,19,20]. LIBs play an essential role in portable electronics, which are the crucial factor in the deployment of electric vehicles due to their high-energy density and long cycle life. Recently, the layered transition metal oxides (Li[NixCoyMnz]O2, nickel cobalt manganese (NCM), or NCMxyz) cathode materials as potential candidates with high-power/energy densities and low cost and toxicity for replacing currently used LiCoO2 in LIBs have been intensively investigated [21, 22]. Among those alternative layered transition metal oxides cathode materials, LiNi1/3Co1/3Mn1/3O2 (NCM) has been hotly studied owing to its superior cycle stability at low cutoff voltage, low cost, and simple preparation. However, there are still many problems that hinder its commercial application [23], such as (1) its high interfacial resistance after cycles and eventual capacity fading owing to the dissolution of transition metals into electrolyte mainly due to the reaction of electrode and electrolyte at the highly delithiated state; (2) its poor rate capability due to low electronic and ionic conductivity; and (3) Li/Ni cation mixing because of the similar radius of Li+ (0.076 nm) and Ni2+ (0.069 nm).

Many effective measures have been used to address these problems. One of the most effective methods is doping with metal or non-metal which can improve the stability of the material structure. Some cations (Na [24], Mg [25], Mo [26], Ti [27], Al [28]) or anions (Cl [29], F [30]) doped in LiNi1/3Co1/3Mn1/3O2 cathode material on transition metal or oxygen sites had been certificated to improve the structure stability and electrochemical activity of LiNi1/3Co1/3Mn1/3O2 cathode materials without participating in the electrochemical reactions. The other effective strategy is surface modification by coating stable materials on electrodes which can protect the electrode surface from direct contact with the electrolyte and suppress the side reactions between the electrolyte and electrode. Wang et al. [31] demonstrated that Li2ZrO3 modification plays an important role to enhance the high-rate capability and cyclability of LiNi1/3Co1/3Mn1/3O2 at various working temperatures. Guo et al. [32] studied the enhancement of the electrochemical performance of LiNi1/3Co1/3Mn1/3O2 by MnO2 modification, which indicated that the discharge capacity of MnO2-coated samples was 155.15 mAh g−1, while that of the pristine sample was 132.84 mAh g−1. Zhang et al. [33] reported that surface modification of LiNi1/3Co1/3Mn1/3O2 with fluoroborate glass resulted in stable cyclability with capacity retention of 96.8% after 50 cycles while the capacity retention of bare one is 81.4%. Recently, some researchers have suggested that “trivalent oxides” (such as La2O3 [34, 35] and Al2O3 [36, 37]) are potential coating materials, compared to other materials, because these coating materials have the potential to prolong the cycle life of Li-ion batteries and surpass the performance of common coatings based on conventional materials [35]. In this work, the electrochemically inactive coating layer La2O3 has been introduced in cathode material, which can effectively prevent surface from interacting with acidic electrolyte components to suppress the side reactions and slow down the TM dissolutions. Though the electrochemical performance of electrodes were improved, the roles of coating materials are difficult to be clarified. Therefore, the roles of the metal oxide coating layer, especially in thin film cathode materials, are worthy of being studied.

In order to deeply research the effect of surface coating on structure and electrochemical performances of cathode material LiNi1/3Co1/3Mn1/3O2, herein, an uncontinuous La2O3-coated LiNi1/3Co1/3Mn1/3O2 has been successfully synthesized via a wet chemical process followed by a solid-state reaction at 800 °C for 5 h. And all the synthesized samples were galvanostatic charged and discharged in the voltage range of 2.5–4.5 V. The La2O3-modified LiNi1/3Mn1/3Co1/3O2 samples show earthshaking improved in cycle performance and rate capacity, which is attributed to the La2O3 layer that reduces electrochemical polarization and suppresses the side reactions between electrode and electrolyte. The structure and electrochemical performances of the La2O3-coated LiNi1/3Mn1/3Co1/3O2 were extensively investigated in this paper.

Experimental section

Materials preparation and characterization

Pristine LiNi1/3Mn1/3Co1/3O2 was provided by the 18th Research Institute Of China Electronics Technology Group Corporation. LiNi1/3Mn1/3Co1/3O2 was coated with the particles La2O3 via a solution-phase method. Stoichiometric amounts of the LiNi1/3Mn1/3Co1/3O2 powder and lanthanum nitrate (La(NO4)3·6H2O) were separately dissolved in absolute alcohol to obtain solution and the mass ratios of La/LiNi1/3Mn1/3Co1/3O2 were 0, 1, 3, and 5 wt% for three different samples, referred to as solution A. Urea as a precipitant was also dissolved in absolute alcohol, referred to as solution B. Then, solution B was gradually dropped into solution A with continuous stirring for 12 h under 80 °C. The precipitated hydroxide powder La(OH)3-coated LiNi1/3Mn1/3Co1/3O2 was filtered off and dried at 120 °C overnight. The obtained powders were calcined at 800 °C for 5 h in air at a rate of 2 °C min−1 to get the La2O3-coated LiNi1/3Mn1/3Co1/3O2. The obtained compound was named as NCM, L-1, L-2, and L-3, respectively. In order to demonstrate that the material prepared via the above preparation process was La2O3 particles, we prepared the LiNi1/3Mn1/3Co1/3O2-free materials according to the above procedure.

The crystal structure of the samples was characterized by a Rigakud/MAX-2500/pc x-ray diffraction (XRD) with Cu Ka radiation at a scanning rate of 5° min−1. The morphological characterizations of the powders were investigated with the Hitachi-S4800 scanning electronic microscope (SEM) and transmission electronic microscope (TEM) with model JEM2010.

Electrochemical measurements

A slurry coating procedure was adopted to prepare the working electrodes. The active material, carbon black as conductive agents, polyvinylidene fluoride (PVDF) as a binder with 80, 10, and 10 wt% mass fraction were dissolved in N-methyl-2-pyrrolidone (NMP) solvent. And then, the slurry was uniformly coated onto the aluminum foil and dried at 120 °C overnight and punched into discs of 10-mm diameter. The mass loading of active materials on the electrode should have an important influence on the electrochemical performances of the designed electrode [38,39,40], In this report, the mass loading of the active cathode materials on the aluminum foil is average 1.8 mg cm−2. The electrochemical performance of the as-prepared sample was evaluated using CR2025-type coin cells with Li electrodes and electrolyte, which were assembled in an argon-filled glove box. The electrolyte employed was a solution of 1 M LiPF6 in a mixture of ethylene carbonate and diethyl carbonate with 1:1 volume ratio. A microporous polypropylene film (Celgard 2400) was used as the separator. The cells were tested by galvanostatic charge-discharge cycling on a battery testing system (LAND, Wuhan, China) at 25 °C. The cyclic test was performed at 1C (1C = 200 mA g−1) with cutoff potentials from 2.50~4.50 V. The rate levels for test were set as 0.2C, 0.5C, 1C, 2C, 5C, 10C, and 0.2C for 10 cycles each with the same potential range as the cyclic test. Electrochemical impedance spectroscopy (EIS) measurements were performed in an alternating current frequency range from 1 mHz to 1 MHz using a CHI660E electrochemical workstation (Chenhua, Shanghai China) in the coin cell.

Result and discussion

The crystal structure of La2O3 powders, NCM, L-1, L-2, and L-3 was characterized by XRD, which can be shown in Fig. 1. From the XRD patterns of the obtained samples in Fig. 1, the diffraction peaks of NCM-free materials prepared by the above process are in accordance with those of La2O3 (JCPDS No. 05-0602), and no extra reflection peak of impurity phase is observed. The x-ray powder diffraction patterns of NCM, L-1, L-2, and L-3 powders exhibited a well-defined layer structure based on a hexagonal a-NaFeO2 structure with an R-3 m space group and no impurity phases due to the low content of La2O3 phase. The distinct splitting of hexagonal doublets (006)/(012) and (018)/(110) peaks indicate an ordered layered structure [41]. There was no significant difference in the crystal structure before and after the coating, indicating the layered structure of LiNi1/3Mn1/3Co1/3O2 remained after the coating. Generally, the cation mixing, which is in consequence of Li/Ni site-exchange because of the similar radius of Li+ (0.076 nm) and Ni2+ (0.069 nm), greatly influences the electrochemical performance, especially the capacity and cycling stability [42]. The Ni2+ ions in the Li layer, which decreased the d-spacing of Li layer in the ab planes due to the higher binding energy between the cation Ni and the anion O atom than between the cation Li and the anion O atom, not only decreased discharge capacity but also impeded to Li+ ionic diffusivity [43, 44]. Moreover, the intensity ratio of I(003)/I(104) is a reliable indicator for determining the cation distribution in the lattice of layered oxides [45, 46], and the value of 1.2 is supposed to be the boundary of cation disordering, and the higher value means better ordering for the cations. The value of I(003)/I(104) for NCM, L-1, L-2, and L-3 samples is 1.138, 1.408, 1.521, and 1.414, respectively. This result indicates that the L-2 sample exhibits the best layered structure and the lowest cation mixing which can facilitate the diffusion of lithium ions. Therefore, the lower cation mixing in La2O3-coated LiNi1/3Mn1/3Co1/3O2 samples infer a better electrochemical performance.

Fig. 1
figure 1

XRD patterns of NCM, L-1, L-2, L-3, and La2O3

Full size image

To explore the influence of the coatings on surface morphology, SEM and TEM were used to analyze the surface morphologies and microstructures of bare and the La2O3-coated LiNi1/3Mn1/3Co1/3O2 samples. As shown in Fig. 2a–d, the SEM images exhibit morphology change of NCM (a), L-1 (b), L-2 (c), and L-3 (d) samples. The primary particle size of the bare and coated LiNi1/3Mn1/3Co1/3O2 samples were distributed with a particle size ranging from 1 to 2 μm and displayed a smooth surface and clear boundary, while the surface-coated LiNi1/3Mn1/3Co1/3O2 was distinctly covered by La2O3 particle with a particle size ranging from 10 to 50 nm. It can form an uncontinuous La2O3 coating layer on the surface of the cathode materials due to uneven particle dispersion and small coating amount. As the coating amount increases, the surface of the particles become rough and form a dense coating layer. The elemental distribution of O, Ni, Co, Mn, and La of the L-3 composite was analyzed by EDS mapping. From Fig. 2e, f, it is shown that elements O, Ni, Co, Mn, and La are uniformly distributed in the sample. SEM image and element mapping of L-3 particles (Fig. 2e, f) revealed that La was successfully coated on the surface of LiNi1/3Mn1/3Co1/3O2 cathode material. Figure 3 shows TEM images of the NCM, L-1, L-2, and L-3 samples. The uncoated sample displayed a smooth surface compared to the coated samples. The surface of the coating material is relatively coarse. When the content of coating is too low, the surface coating layer of the material is incomplete. The coating layer appears as small particles adsorbed on the surface of the LiNi1/3Mn1/3Co1/3O2 cathode material. As the coating content increases, the size of particles increases and the coating layer becomes complete and thicker. Thus, from the analyses of the XRD, SEM, TEM, and EDS patterns, we can come to the conclusion that the La2O3 successfully coated on the surface of L-1, L-2, and L-3 samples.

Fig. 2
figure 2

SEM images for a NCM, b L-1, c L-2, and d L-3. e–f EDS mapping images for L-2 sample

Full size image
Fig. 3
figure 3

TEM images for a NCM, b L-1, c L-2, and d L-3

Full size image

The galvanostatic charge-discharge tests have been performed to compare the electrochemical performance of pristine LiNi1/3Co1/3Mn1/3O2 and La2O3-modified LiNi1/3Co1/3Mn1/3O2. The charge-discharge capacity profiles at different cycles (1st, 10th, 20th, 30th, 40th, 50th, 60th, 70th, 80th, 90th, and 100th) of the pristine and L-2 electrode at 1.0C in the voltage range of 2.5–4.5 V versus Li are shown in Fig. 4a, b, respectively. There is only one voltage plateau at 3.5–3.8 V observed for the as-synthesized materials in the cycle corresponding to the redox activity of the transition metal cations, namely, Ni, with typical layer-structured LiNi1/3Co1/3Mn1/3O2. Before the rapid drop at about 3.5 V in the discharge curves, voltage decrease of the L-2 electrode (in Fig. 4b) is quite smooth compared with the pristine electrode (in Fig. 4a), and the platforms of charge and discharge curves for L-2 electrode are longer than pristine electrode, indicating a higher energy density and a great cycling stability. Notably, the pristine electrode suffers from observable capacity/voltage fading. The dissolution of transition metal ions might take place upon long cycles, along with the crystal destruction and the decomposition of electrolyte, leading to severe performance decay. Figure 4c shows the specific capacity and Coulombic efficiency of the NCM, L-1, L-2, and L-3 electrodes at 1C rate between 2.5 and 4.5 V are plotted versus cycle number. The cycling data of the electrodes are listed in Table 1. A significant difference was found in the field of the capacity retention; the capacity retention of cathodes was hugely improved after surface modification. The initial specific capacities of the NCM, L-1, L-2, and L-3 electrodes were found to be 181.6, 181.0, 181.3, and 179.9 mA h g−1, respectively. And the capacity retentions of the NCM, L-1, L-2, and L-3 electrodes are 38.6, 87.6, 96.1, and 87.1% over 100 cycles, respectively. Obviously, surface coating of La2O3 can greatly enhance the cycle performance of the LiNi1/3Mn1/3Co1/3O2 material. The L-2 electrode possesses the highest discharge capacity after 100 cycles and that means the best cycling stability for the modified sample. The good cycling stability of the surface-modified samples was due to the presence of coating layer, which protected the bulk material from being corroded by electrolyte and suppressed the electrolyte decomposition during cycling. The IR-drops for the NCM and L-2 electrodes of the 1st, 10th, 20th, 30th, 40th, 50th, 60th, 70th, 80th, 90th, and 100th cycle are presented in Fig. 4d. It is known that LIB cathodes suffer from the effect of polarization, which leads to the open-circuit voltage decay and thus the starting discharge voltage is usually lower than 4.5 V [47]. The reducing value in voltage is a criterion for evaluating the effect of polarization [48]. As can be seen from the IR drop in Fig. 4d, the pristine electrode shows an initial IR drop from 4.5 to 4.38 V, which then decrease to 4.04 V after 100 cycles gradually. In comparison, the IR drop of L-2 electrode is almost stabilized at 4.4 V in 100 cycles. This implies a larger polarization for the pristine electrode. The energy densities and average discharge voltage (illustration) versus cycle number plots at 1C of NMC and L-2 materials are shown in Fig. 4f and Table 1. The decrements of the average discharge voltages for L-2 electrode is 10.3 mV over 100th cycle and 26.9 mV over 200th cycle compared to 407 mV over 100th cycle for bare NCM. The continuous phase transformation from the layered to spinel structure during the charge-discharge process may result in the voltage decay, as shown in previous reports [49, 50]. The energy densities of the NCM and L-2 electrode are 688.65 and 694.80 Wh kg−1, respectively. And the energy densities retention of L-2 is 96.5% after 100 cycles and 89.7% after 200 cycles compared to 34.5% after 100 cycles for bare NCM. The L-2 electrode displays capacity and voltage stability obviously better than those of the NCM electrode. All these results indicate that La2O3 coating on NCM material can effectively reduce the polarization of the material, enhance cycle stability, and effectively suppress voltage and energy attenuation.

Fig. 4
figure 4

The charge-discharge curves of the NCM (a) and L-2 (b) electrodes. c Plots of the discharge capacity with the Coulombic efficiency versus cycle number for as-prepared electrodes. d The initial discharge voltage of the NCM and L-2 electrodes (IR drop). f Energy densities and average discharge voltage (illustration) versus cycle number plots at 1 C of NMC and L-2 materials. Those electrodes measured at a 1C rate in the potential range of 2.5–4.5 V versus Li

Full size image
Table 1 Cycling performance of the NCM, L-1, L-2, and L-3 electrodes
Full size table

To further evaluate the electrochemical properties of the as-prepared cathodes, the rate capabilities were tested in the range of 2.5–4.5 V at various current densities and the discharge curves at various rates, as shown in Fig. 5. As expected, the specific capacities of all cathodes decrease with an increase in the current density due to the increasing polarization and L-2 electrode exhibits a best rate capability than other electrodes. The discharge capacity of NCM cathode material is 189.0, 172.7, 158.4, 140.3, 115.8, and 86.9 mAh g−1 at 0.2C, 0.5C, 1C, 2C, 3C, and 5C, respectively. While the L-2 cathode material offers the discharge capacity of about 187.0, 182.0, 171.2, 157.2, 146.8, and 137.5 mAh g−1 at the same rates, respectively. From Fig. 5a, it can be seen that when the current density was increased to high rate, the significant differences can be observed clearly. The pristine NCM shows severe capacity degradation as the rate increases from 0.2C to 5C. In comparison, L-2 electrode exhibits much better rate capability, suggesting better structural stability of the coated sample for rapid Li+ insertion/extraction. Figure 5b, c show the discharge curves at different C rates. One of the most important reasons, which result in dramatically reduction of the working voltage at rates of 0.2C and 5C for the pristine NCM, is the undesirable increased interphase resistances caused by the side reaction on the interface between electrode and electrolytes [51]. In comparison, the voltage fading and capacity loss are obviously restrained in L-2 electrode.

Fig. 5
figure 5

a Rate capability tests at several currents (C rates) for NCM, L-1, L-2, and L-3 electrodes in the potential range of 2.5–4.5 V versus Li. Discharge profiles of NCM (b) and L-2 (c) electrodes cathode at various C rates of 0.2, 0.5, 1, 2, 3, and 5

Full size image

To further analyze the electrochemical nature of pristine and L-2 electrodes, CV measurements were conducted at voltages between 2.5 and 4.5 V at a scan rate of 0.1 mV s−1. The CV curves of all electrodes after 100th cycle are illustrated in Fig. 6. There is a pair of oxidation and reduction peaks in every curve, which corresponds to Ni2+/Ni4+ redox couple [52]. As shown in Fig. 6, their anodic peaks are 3.908 and 3.842 V, reduction peaks are 3.681 and 3.698 V, and their redox reaction gaps can be calculated to be 0.23 and 0.144 V, respectively, indicating that the polarization of the sample L-2 is small compared with the pristine samples. The above results suggest that the sample L-2 has the best electrochemical performance than others.

Fig. 6
figure 6

Cyclic voltammetry data for NCM and L-2 electrode after 100th cycle and voltage range from 2.5 to 4.5 V

Full size image

Figure 7a shows the Nyquist plots of NCM and L-2 electrode after 100th cycling. The Nyquist plots are fitted and presented in Fig. 7b, in the medium frequency region, a semicircle can be found, which is related to the charge-transfer resistance of the electrode. A line at low-frequency region with about 45° to the real axis is assigned to the Li+ diffusion within the electrode [53]. In the equivalent circuits inserted in Fig. 7b, Rs represents the solution resistance, which includes the ohmic resistance in NCM or L-2 materials and the tab of cathode/anode. Re represents the diffusion resistance of Li+ in the surface layer (including SEI film). The parameter Rct corresponds to the charge-transfer resistance, CPE1 refer to the nonideal capacitance of the surface layer, and CPE2 is assigned to charge-transfer capacitance, Zw refers to the Warburg impedance, which represents the diffusion behavior of Li+ in the bulk [54]. The simulated values after the 100th cycle are shown in Table 2; the resistances of L-2 are lower than that of NCM, implying a better high rate performance in L-2 than NCM. From Fig. 7, it can be seen that La2O3-coated LiNi1/3Mn1/3Co1/3O2 material could form an SEI film to protect the NCM electrode after 100 cycles. The Rct of NCM and L-2 according to the fitting results were 149.6 and 54.01 Ω, respectively, indicating the improved electronic contact and more effective charge transport of L-2 compared with NCM, which was in accordance with difference of the sizes of the two semicircles. It indicates that a protective layer has been formed on the surface of the cathode materials during cycles, which can reduce the degradation of material structure and effectively improved the electrochemical kinetics of the material.

Fig. 7
figure 7

a Fitted Nyquist plots for NCM and L-2 electrode after 100th cycle. b The corresponding equivalent circuits for NCM and L-2 electrode

Full size image
Table 2 Electrochemical impedance of the NCM and L-2 electrodes
Full size table

Conclusion

High-voltage layered LiNi1/3Mn1/3Co1/3O2 oxides, which can deliver high-energy density, are highly desired for next-generation energy storage devices and capable of addressing some of the challenges associated. However, the cathode material suffers from severe voltage and capacity attenuation at high cutoff voltage due to the structure instability and side reaction between electrode and electrolyte at the highly delithiated state. To improve the high voltage stability and high-energy density of the cathode material, an uncontinuous La2O3-coated LiNi1/3Mn1/3Co1/3O2 has been synthesized via a wet chemical process followed by a solid state reaction at 800 °C for 5 h. It can be observed from the XRD patterns and SEM and TEM images that the La2O3-coated LiNi1/3Mn1/3Co1/3O2 samples with uncontinuous La2O3 coating showed no change in the structure and morphology as compared to pristine NCM. La2O3-coated LiNi1/3Mn1/3Co1/3O2 samples show greatly elevated discharge capacity, energy density, and rate capacity due to the protective role of the La2O3 layer, which suppresses the side reactions between electrode and electrolyte, effectively slows down the voltage decay, and reduces electrochemical polarization during cycling of LiNi1/3Mn1/3Co1/3O2. Through the surface modification method, 3 wt% La2O3-coated LiNi1/3Mn1/3Co1/3O2 illustrated highest rate capability, lowest voltage decay, outstanding cycling performance, and excellent energy density, which attribute to a protective layer that has been formed on the surface of the cathode materials during the cycles. The discharge capacity of NCM at a 5C rate increases from 86.9 to 137.5 mA h g−1 upon coating with La2O3. Furthermore, the layer stability of LiNi1/3Mn1/3Co1/3O2 has been enhanced after La2O3 coating, which results in improved midpoint voltages. The capacity retention of 3 wt% La2O3-coated LiNi1/3Mn1/3Co1/3O2 is 96.1% after 100 cycles compared to 38.6% after 100 cycles for bare NCM. The decrements of the average discharge voltages for 3 wt% La2O3-coated LiNi1/3Mn1/3Co1/3O2 electrode is 10.3 mV over 100th cycle and 26.9 mV over 200th cycles compared to 407 mV over 100th cycle for bare NCM. And the energy densities retention of 3 wt% La2O3-coated LiNi1/3Mn1/3Co1/3O2 electrode is 96.5% after 100 cycles and 89.7% after 200 cycles compared to 34.5% after 100 cycles for bare NCM. All the above results show that the surface coating of LiNi1/3Co1/3Mn1/3O2 with La2O3 is an effective way to improve its electrochemical performance.