Unraveling the structural transformation of Li-rich materials in lithium-batteries

Lithium-Ion Batteries (LIBs) are essentials in everyday life in mobile applications as well as in hybrid/electric mobility. The extraordinary market success of this technology is forcing hard the need of LIBs with improved energy density, environmental compatibility and safety, making necessary to push this technology beyond the current state of the art. In this framework, Co-poor Lithium Rich Layered Oxides (LRLOs) are the most strategic alternative to current Co-rich layered oxide positive electrode materials thanks to the excellent combination of large specific capacity (>250 mAhg^-1), high energy density (up to 900 WhKg^-1), small costs and improved environmental benignity. The excellent performance of LRLOs derives from the peculiar combination of redox processes originated from the transition metals and the oxygen anions sublattice. The practical use of LRLOs is hindered by several drawbacks, such as voltage decay, capacity fading, and an irreversible capacity lost in the first cycle. These issues are related to structural rearrangements in the lattice upon cycling.

In this work, we demonstrate a new family of LRLOs with general formula Li_1.2+xMn_0.54Ni_0.13Co_x-yAl_0.03O₂ (0.03 ≤ x ≤ 0.08 and 0.03 ≤ y ≤ 0.05), obtained from the replacement of cobalt with lithium and aluminum and we highlight how the balancing of the metal blend can lead to improvements of the Coulombic efficiency in the first cycle, a better capacity retention and reduced voltage decay. To shed light on the complex crystal-chemistry of this class of LRLOs we studied the Co-poorest member of this homologue material series, namely Li_1.28Mn_0.54Ni_0.13Co_0.02Al_0.03O₂, in order to prove the structural evolution occurring upon charge/discharge in lithium cell. To this aims, electrodes have been recovered during the first cycle, the second cycle and after ten cycles of charge/discharge by de-assembling lithium cells into an Ar-filled glove box. These post mortem materials have been sealed in borosilicate capillary tubes (see Fig. 1a) and studied ex situ by X-ray powder diffraction at the MCX beamline.

Figure 1  a) sketch of procedure used to prepare ex-situ samples; b) cell potential vs time during the first two charge-discharge cycles; c) diffractograms collected during the first cycle (indicated by the labels) highlight the changes in the lattice parameters.

Fig. 1b shows the potential curves vs time for the first two cycles and highlights the points, marked with A, B, C and so on, where the charge or discharge step was stopped and the materials recovered for analysis. According to the diffraction data (Fig. 1c), structural alterations of Li_1.28Mn_0.54Ni_0.13Co_0.02Al_0.03O₂ start with a fast broadening and a shift of the peaks suggesting a smooth lattice modification. When the cell reaches 4.8V vs Li⁺, a second phase can be identified. In the discharge process opposite structural transformations occur.

Figure 2: Rietveld refinement determiation of the cell parameter c for (a) the first and (b) the second cycle.

The high-resolution patterns have been analysed by Rietveld refinement, assuming a rhombohedral unit cell, to disclose quantitative information about the lattice parameters and their changes. The main results are reported in Fig. 2a-b for the first and the second cycle in battery. The changes in the a and c parameters of the rhombohedral unit cell are highly correlated with the redox processes occurring at different voltages. The shrinking of the a parameter is related to the oxidation of Ni²⁺ and Co³⁺ to Ni⁴⁺ and Co⁴⁺. Overall, the oxidation of transition metals leads to more compact crystal packing thank to the decrease of ionic radii and the increase of the net charges. These phenomena result in the shortening of the M-O bond length and in the shrinking of the MO₆ octahedra. On the other hand, the c parameter increases due to loosening of the staking among the O-Li-O-M layers due to the removal of lithium ions and the consequent increase of the electrostatic repulsions between vicinal oxygen layers. The opposite trend is found during the discharge mode. Remarkably, the Rietveld refinement analysis confirms the segregation of a second phase at the end of the first charge (at 4.8V, corresponding to x=0.8 in Fig. 2a). This new phase is isostructural to the parent pristine material and, once formed, does not disappear upon cycling. Both phases participate in the reversible electrochemical reactions, as demonstrated by the changes in the lattice parameters (Fig. 2b).

The development of novel materials for next generation batteries requires to carefully balance performance improvements, enhanced environmental benignity and cost sustainability along the entire life-cycle, from synthesis to recycling. This study exploits the use of a synchrotron-based technique to characterize the working mechanism of LRLOs in batteries: this is a crucial step to promote a knowledge-based design of innovative materials for high-capacity positive electrodes.

This research was conducted by the following research team:

Arcangelo Celeste^1,2, Fabio Girardi³, Lara Gigli⁴, Vittorio Pellegrini^1,5, Laura Silvestri⁶ and Sergio Brutti⁷

¹Istituto Italiano di Tecnologia, Genova, Italy
²Dipartimento di Chimica e Chimica Industriale, Università degli Studi di Genova, Genova, Italy
³Dipartimento di Fusione e Tecnologie per la Sicurezza Nucleare, ENEA C.R. Casaccia, Roma, Italy
⁴Elettra-Sincrotrone Trieste, Basovizza, Trieste Italy
⁵BeDimensional Spa, Genova, Italy
⁶Dipartimento di Tecnologie Energetiche e Fonti Rinnovabili, ENEA C.R. Casaccia, Roma, Italy
⁷Dipartimento di Chimica, Sapienza Università di Roma, Roma, Italy

Contact person:

Local contact person: (MCX beamline)