X-ray diffraction analysis
Fig 2 displays the XRD patterns for pristine LiNi1/3Mn1/3Co1/3O2 and RuO2 doped LiNi1/3Mn1/3Co1/3O2. As seen in Fig 1a, the peak positioned at 2θ = 18.7, 44.5, 49.5, 36.6 and 66.2 are consistent with (003), (104), (105), (101) and (108) crystal planes of LiNi1/3Mn1/3Co1/3O2 and the preferred orientation is being at (104) plane. But for the other three RuO2 doped material which exhibited the similar XRD pattern with enhanced peak intensities suggested the existence of better crystal structure. A weak peak diffracted at 2 theta value of 28.2 is due to the presence of RuO2 on the surface of LiNi1/3Mn1/3Co1/3O2 is shown in Fig 1(b-d). However, the weak intense peaks at 43.20 and 63.40 indicate the cubic structured phase of LiNiO2. For all the samples, the diffraction peaks were sharp and well definite revealed the good crystallinity and the crystal structure was recognized as a hexagonal α-NaFeO2 structure with R3m space group [25,26] that approves the occurrence of layers of Li, Ni, Mn, and Co in a single-phase layered structure. 3wt % RuO2 doped sample exhibited higher intensity crystal planes than that of other samples indicating the importance of RuO2 and good crystallinity.
Surface morphology analysis
SEM characterization was used to study the morphology of pristine LiNi1/3Mn1/3Co1/3O2 and RuO2 doped LiNi1/3Mn1/3Co1/3O2 samples and are shown in Fig 3. All the samples presented in the SEM images are well-crystalline polyhedral nanoparticles with a size of about approximately 200 nm. As seen in Fig 3a, pristine LiNi1/3Mn1/3Co1/3O2 showed a layered structure with high uniformity and is in good agreement with the previous reports [27]. In the case of 3 wt% RuO2 doped samples, RuO2 nanoparticles are entrapped on the surface of pristine LiNi1/3Mn1/3Co1/3O2 that may be due to the high surface energy of the nanoparticles forming a surface doped layer which is shown as in Fig 3b. On increasing the concentration of RuO2 (3 %), the coatings become more compact. It is clear that, all the RuO2 doped samples showed a rough surface than that of pristine material. This rough surface may enhance the electronic conductivity of the Li-rich material, lowering the ion diffusion resistance, providing a stable protective structure for the bulk material, and avoiding the occurrence of surface side reactions between the electrode material and electrolyte.
In order to get further insight about the surface morphology, TEM analysis was carried out. Fig 4a shows the formation of similar polyhedral morphology as observed in SEM picture. The introduction of 3 % of RuO2 does not show any significant morphological changes but the embedded nanoparticles are clearly seen in Fig 4b.The crystalline size of the nanoparticles is estimated as approximately 200 nm. The corresponding SAED pattern shows the ring pattern along with bright spots suggested the poly crystalline nature of the proposed battery material is shown in Fig 4c. Lattice fringes are also well agreed with the crystal planes and are shown as Fig. 4d. The wider fringe spacing is 0.47 nm for the (003) planes of the layered structure, and the slightly finer lattice fringe is 0.30 nm for the (111) planes of the rutile structured RuO2. However, the fringe spacing of RuO2 is lower than that of pure RuO2 crystals (0.31 nm), which is ascribed to the doped Ru element.
EDX analysis
The chemical composition of RuO2 doped LiNi1/3Mn1/3Co1/3O2 was studied from energy dispersive X-ray spectroscopy (EDX) Fig. 5. The EDX spectrum reveals the existence of Ni, Mn, Co O and Ru elements in the Li-rich material with the wt% of 18.6, 19.4, 25.6, 31.1 and 8.1 respectively. Inset shows the actual weight % of the elements such as Li, Ni, Co, Mn, and O along with Ru indicating that the surface of LiNi0.23 Ru0.1Mn0.33Co0.33O2 is decorated with RuO2 nanoparticles.
XPS analysis
The XPS technique was used to investigate the elemental composition of the as-prepared material and the valence states of elements (Fig 6). However, it can be found that the orbital binding energy of Ni 2p3/2 is located at 851 eV in Fig. 6a which significantly shifts to higher binding energy positions (856 eV) indicating that a part of Ni2+ turns into higher valence state. According to this result, it could be concluded that partial Ru4+ ions enter the crystalline lattice of the LiNi1/3Mn1/3Co1/3O2 material and replace a part of Ni2+ ions. From the XPS spectra of Ru 3p, it can be seen that the main peaks at about 465.3, 462.8 eV correspond to Ru 3p5/2 and Ru 3p3/2 signals of Ru4+ and Ru3+ respectively [28] with no significant chemical shift. The addition of Ru promotes the generation of more number of oxygen vacancies. In addition, it also facilitates the redox reaction between Ru4+ and Ru3+ of RuO2 and further enhances its oxygen storage capacity. The two main peaks centered at 781.1 eV and 797.2 eV are due to the 2p3/2, 2p1/2 spin orbit splitting of cobalt respectively [29]. The more oxygen vacancies were also verified using O 1s spectra of the as-prepared materials. In view of the analysis, the peaks located at 529.2 and 531.2 eV were assigned to lattice oxygen, oxygen vacancies, and chemisorbed oxygen, respectively [30]. Obviously, the content of oxygen vacancies is further increased after the incorporation of Ru. It is clear that Ru doped LiNi1/3Mn1/3Co1/3O2 possesses more oxygen vacancies, which is favorable for promoting the activation of LiNi1/3Mn1/3Co1/3O2. From the above results, it can be concluded that Ru4+ ions incorporated into the pristine material was expected to enhance the electrochemical performance of the material.
Electrochemical Impedance spectroscopy
Electrochemical Impedance spectroscopy is used to study the kinetics during lithium intercalation/deintercalation process. Fig 7 shows the electrochemical impedance spectroscopy (EIS) profile of pristine LiNi1/3Mn1/3Co1/3O2 and RuO2 doped LiNi1/3Mn1/3Co1/3O2. The impedance spectrum consists of a semicircle in the intermediate frequency ranges followed by an inclined straight line at the low frequency range. In general, the semicircle in the high to medium frequency region is related to the charge transfer resistance (Rct) [31,32] and in the low frequency region represents the Warburg impedance (Zw), which is ascribed to Li-ion diffusion in the solid phase state of the electrode material. According to Fig 7, the Rct value of the RuO2 doped electrode is smaller than that of the un doped pristine electrode. The diameter of the semicircle for the un doped electrode is 226.79 Ω and that of the doped electrode is 205.47 Ω. Since the diameter decreases for the RuO2 doped electrode has confirmed the enhancement of conductivity. As a consequence, the electrochemical properties get improved. The equivalent circuit is shown as inset in Fig 7.
Electrochemical performance of RuO2 doped LiNi1/3Mn1/3Co1/3O2 cathode material
The electrochemical performance of the proposed battery material was investigated by galvanostatic charge/discharge curves at 0.1 C rates in the voltage range between 2.8 to 4.2 V. As seen in Fig 8, the charge/discharge curve plateaus meets at around 3.9 V and the potential drop was observed at 3.6 V warrants the higher energy density. In general, the wider charge/discharge profile suggests the better stability of the electrode. The discharge capacity for the pristine material was estimated as 194.9 mAhg-1 whereas for 1, 2 and 3 wt % of RuO2 doped LiNi1/3Mn1/3Co1/3O2, the discharge capacity values are calculated as 214.9, 242.9, and 251.2 mAhg-1 respectively. As expected, 3 wt % RuO2 doped sample exhibits better cyclic performance compared to other electrodes which is well agreed with the literature [33]. The embedded RuO2 particles suppress the metal ion dissolution and unwanted side reaction between the electrode and electrolyte leads to the enhanced Li transportation. Consequently, rate performance of the proposed cathode material was improved. After 100 cycles, 3.2 % capacity fade was observed for 3% RuO2 doped sample which may be due to the dissolution of transition metal ions at higher voltage or the electrolyte decomposition.
In order to test the electrode stability and capacity retention, galvanostatic charge /discharge test was carried out (Fig 9). All the electrodes were charged and discharged at 0.1 C rates between 2.8 to 4.2 V for 100 cycles. For the pristine LiNi1/3Mn1/3Co1/3O2, the initial discharge capacity is 194.9 mAhg-1, and after 100 cycles the discharge capacity is reduced to 143.27mAhg-1 and 74 % of capacity was retained. In the case of 1% of RuO2 doped LiNi1/3Mn1/3Co1/3O2, 75 % capacity retention was obtained whereas for 2% RuO2 doped material capacity remains as 217 mAhg-1 and the retention is 89 %. 3% RuO2 doped LiNi1/3Mn1/3Co1/3O2, shown better capacity retention and long lasting stability. The initial discharge capacity is 251.2 mAhg-1 and it was slightly declined to 226 mAhg-1 after 100 cycles indicating the superior performance of the proposed cathode material. The doping strategy revealed the better performance than the previous reports [34]. The capacity retention was determined as 90 %. The addition of RuO2 facilitates the transportation of Li2+ ions in the electrode-electrolyte interface. The existence of Ru in crystal lattice could enhance the ability of Li diffusion, more Li ions can be easily extracted from Li layers. Consequently, higher extent of phase transformation from layered Li2MnO3 to certain spinel-like regions could be expected. As a consequence, capacity caused by reinsertion of Li ions into local transformed spinel-like regions during discharge will be increased. Another reason for the higher capacity contribution from spinel-like regions is that the appropriate amount of Ru (x = 0.01) gives rise to a certain impact on transferred spinel-like lattice to enhance Li diffusivity in these regions [35]. It is worth mentioning that, on increasing the cycle numbers the specific capacity also increases owing to the good activation of the electrodes.
Cyclic Voltammetry is a useful method for determining the structural changes in Lithium intercalation/deintercalation process. To get further information about the electrochemical characteristics, Cyclic Voltammetry experiments were performed in the scan range of 2.8 to 4.2 V at a scan rate of 1 mVs-1. Cyclic voltammograms of first six cycles of RuO2 doped LiNi1/3Mn1/3Co1/3O2 is shown in Fig 10. A double layer capacitive behavior was observed rather than the typical redox behavior of RuO2.