Unraveling the effect of exposed facets on voltage decay and capacity fading of Li-rich layered oxides
Graphical abstract
Voltage decay and capacity fading of Li-rich layered oxides can be mitigated by large area of exposed inactive surface.
Introduction
The mass application of Li-ion batteries in the electric vehicle and large-scale energy storage is greatly limited by the availability of high-energy density cathodes. Therefore, the development of cathodes with high performance is urgently needed [1]. In this regard, Li-rich layered oxides (LLOs, xLi2MnO3·(1-x)LiMO2 or Li1+xM1-xO2, M = Mn, Co, Ni) have been widely investigated due to their high specific capacity (ca. 250 mA h g−1) and low cost [2], [3]. The high specific capacity of LLOs originates from the electrochemical activation of Li2MnO3 component over 4.5 V [4], [5]. However, the activation of Li2MnO3 component leads to high irreversible capacity loss and surface rearrangement of LLOs [6]. In LLO cathodes, Li-ions can only transport along the plane groups paralleled to the (001) facet [7]. It makes many characters of LLOs, such as rate capability and element segregations, have a close relation with exposed facets.
Up to now, literature have demonstrated that high percentage of active exposed planes, which have unimpeded pathways for Li-ions transportation, can help to optimize the rate capability of LLOs [8], [9]. For instance, Wu et al. and Sun's group have shown that the rate capability of LLO nanoplates can be greatly improved by increasing the percentage of {010} active exposed planes [10], [11]. In addition, Wang's group and Nanda et al. have found that both element segregations and phase transformation of LLOs are facet-dependent [12], [13], [14]. For element segregations and phase transformation, they are the most undesirable phenomena since they will result in voltage decay and capacity fading of LLOs [15], [16], [17]. However, the effect of inactive and active facets on element segregations and phase transformation of LLOs is far from clear. This is because LLOs with exposed inactive or active planes used to do the above study should be prepared by very similar method to ensure a single variable. It greatly enhances the difficulty of the study on the role of inactive and active facets in element segregations and phase transformation of LLOs.
In this work, two nanoplates are prepared by very similar co-precipitated-precursor method. The top-bottom surface of one nanoplate is (101) facet (S101 sample, active plane) and the other is exposed (001) facet (S001 sample, inactive plane). Electrochemical test suggests the rate capability of S101 sample is better than that of S001 sample. It is due to the active exposed plane which can shorten the transport distance of Li-ions in electrode materials. However, both the capacity and voltage of S101 sample decrease much faster than those of S001 sample. This is because large area of inactive surface can mitigate electrochemical corrosion and phase transformation of LLOs in terms of the discussion of TEM, HRTEM, XPS and XRD for cycled samples. In addition, we discover Co and Ni are easier to react with electrolyte than Mn.
Section snippets
Sample synthesis
Two LLO nanoplates were prepared by very similar co-precipitated-precursor method. Analytical grade chemicals of nickel sulfate (NiSO4·6H2O), manganese sulfate (MnSO4·H2O), cobalt sulfate (CoSO4·7H2O), sodium hydroxide (NaOH), diglycol (NMP: N-methyl pyrrolidinone), and ammonium hydroxide (NH4OH) were chosen as the starting materials. For synthesis of the nanoplate with exposed (101) plane, 1825.4 mg MnSO4·H2O, 1209.1 mg NiSO4·6H2O, 1293.1 mg CoSO4·7H2O, and 50 mL NMP were dissolved in 100 mL
Structure and morphology of the nanoplates
The crystal structure of the as-prepared nanoplates is analyzed by XRD, as shown in Fig. 1. All the strong diffraction peaks for the two XRD patterns can be indexed to a typical hexagonal α-NaFeO2-type structure with a space group R-3m [18], except for the weak peaks in the 2θ range of 20–25°. These weak peaks are caused by lithium-cation ordering in the transition-metal layers. It is the features of the integrated monoclinic Li2MnO3 phase (C2/m) [19]. In addtion, both the pair reflections
Conclusions
In this work, two nanoplates are prepared by very similar co-precipitated-precursor method. The top-bottom surface of one nanoplate is (101) facet (S101 sample) and the other is exposed (001) facet (S001 sample). The two nanoplates have similar chemical composition, particle size, and surface oxides etc., and their difference is mainly concentrated in exposed facet. This difference results in the following facts: the rate capability of S101 sample is better than that of S001 sample, and the
Acknowledgments
This work is financially supported by the National Natural Science Foundation of China (No. 21373136) and the Natural Science Foundation of Shanghai (No. 16ZR1416800).
References (37)
- et al.
Electrochim. Acta
(2015) - et al.
Electrochem. Commun.
(2006) - et al.
J. Power Sources
(2011) - et al.
Nano Energy
(2016) - et al.
Electrochim. Acta
(2014) - et al.
J. Power Sources
(2002) - et al.
Chem. Mat.
(2012) - et al.
Nat. Mat.
(2015) - et al.
ACS Appl. Mat. interfaces
(2014) - et al.
Nano Lett.
(2013)