Transition Mechanism and Phase Transition Front of LixNi0.5Mn1.5O4

1. Introduction

Nickel substituted manganese spinel LiNi_0.5Mn_1.5O₄ (LNMO) shows high rate capability even it consist of micron-sized particles.^[1] This is in contrast to LiFePO₄ where only nano-sized particles show high rate performance. For the designing of high performance batteries, it is important to understand the phase transition mechanism of the micron-sized LNMO that is inseparably connected with the rate capability.^[2]

As the micron-sized LNMO particles is expected that the existence of possible intermediate states between the phases (namely Li1, Li1/2 and Li0) are captured by operando XRD while TEM analysis shows the existence of the phase transition front. In this study, we revealed the factor of reaction kinetics from phase transition behavior of the LNMO by operando (1 C and 5 C) synchrotron X-ray diffraction (XRD) measurement, and transmission electron microscopy (TEM) analysis.

2. Experimental procedure

LNMO powder used as an active material has primary and secondary particle sizes of around 1 and 5-10 µm, respectively. The working electrode consisted of LNMO (80 wt.%), acetylene black (10 wt.%) and polyvinylidene difluoride (PVdF) binder (10 wt.%), coated on aluminum current foil.

The electrochemical measurements of the electrode were employed using aluminum pouch type cells, metallic lithium as counter and reference electrodes, 1 mol dm^-3 LiPF₆ in a 3:7 mixture of EC/DMC, and a polyolefin film as a separator. The electrochemical charge and discharge tests were performed at room temperature and the potential range was between 3.5 and 5.0 V.

The time-resolved XRD (TR-XRD) measurements were conducted at the BL28XU and BL46XU of SPring-8, Hyogo, Japan. The incident X-ray of 12.4 keV (0.100 nm wavelength) was used. We measured the diffractions in the region around the 115 peak of LNMO. The data acquisition time was 0.5 s and discharge tests were employed at rate of 1 or 5 C.

The TEM was employed to analyze the particle morphology of initial LiNi_0.5Mn_1.5O₄ powder and Li_0.25Ni_0.5Mn_1.5O₄ obtained by 1 C charging. Electron diffraction (ED) analysis and electron energy-loss spectroscopy (EELS) were also performed to analyze crystal structure and intraparticle distribution of the reacting material.

3. Results and Discussion

Fig. 1 shows the XRD patterns obtained during the transition between Li1 and Li1/2, and Li1/2 and Li0 at charging process. When two symmetric peak profiles using the Gaussian functions are assumed for these phases at a region of two phases existence, the experimentally obtained intensity in between the two peaks is larger than the calculated intensity. It’s suggesting that there are minor diffractions with intermediate d values in addition to major ones of the phases. A good fit is obtained assuming two additional diffraction components, implying the existence of intermediates in between the phases.

Compared to the homogeneous morphology of the pristine LNMO particle, cracks, stripes, lattice defects and regions of different contrast are observed in the charged sample as shown in Fig. 2. With ex-situ TEM analysis the coexistence of the two phases in primary particles is shown, suggesting that the phase front movement is slow and that the minor diffractions observed in operando XRD are originated from solid-solution domains at the phase transition front of Li_xNi_0.5Mn_1.5O₄. At 5 C rate single phase transition behavior is observed for both Li1 and Li0 phases.

 References

[1] X. Ma, et al., J. Electrochem. Soc., 2010, 157, A925.

[2] T. Ichitsubo et al., J. Mater. Chem. A, 2013, 1, 14532.

[3] H. Arai et al., J. Mater. Chem. A, 2014, 2, 15414.

Acknowledgement

This work was supported by RISING of NEDO.

Figure captions

Fig. 1 XRD profiles of a LiNi_0.5Mn_1.5O₄ electrode in 1 C charging. Two two-phase coexistence regions are fitted with four Gaussian functions.

Fig. 2 TEM image and ED pattern of (a) LiNi_0.5Mn_1.5O₄(pristine) and (b) Li_0.25Ni_0.5Mn_1.5O₄ electrode sample obtained by 1 C rate charging.