Coherent Phase Transformation in Next-Generation Nanoscale Olivine Cathodes

Due to its outstanding power, safety and cycle-life olivine LiFePO₄ (LFP) has during the past decade become a widely used, and one of the most well-studied, lithium ion battery cathode materials. It is well-established that for LiFePO₄ the storage/release of lithium is accompanied by a first-order phase transition between lithiated and delithiated states. First-order phase transition behavior persists well into the nanosize regime, although increasing mutual miscibility is observed with decreasing particle size.^1-4 Consequently, there have been numerous attempt to describe the detailed mechanism of the phase transformation in nano-LFP. Recently, direct observations by oxidation-state-sensitive X-ray microscopy combined with transmission electron microscopy, have been used to show that at intermediate states of charge, nanoparticulate LFP electrodes are primarily a binary mixture of lithiated and delithiated with relative absence of two-phase particles.⁵ This preference has important implications for the kinetics of storage in the large arrays of nanoscale particles that comprise practical battery electrodes.

However, it would be a mistake to conclude that the behavior of pure LFP is representative of all olivines, in particular the vast range of doped and mixed-metal olivines that are also of interest for their advantageous electrochemical properties.^4,6 This is illustrated herein through investigation of the phase transformation in nanoscale olivine LiMn_yFe_1-yPO₄ (LMFP) during dynamic battery operation.

LMFP may be regarded as the second generation olivine cathode, as it has both higher energy density than LFP due to higher average voltage, and lesser known, among the highest rate capabilities ever observed for olivine cathodes. It has been suggested⁷ that this is due to the existence of an intermediate solid solution (here denoted L_xMFP) that reduces the misfit strain by breaking the single phase transition (observed for LFP) into two stages, first a transition of MFP to an intermediate phase L_xMFP, then between L_xMFP and LMFP. However, no clear conclusion has been drawn as to whether the transformations are two-phase first-order reactions or involve formation of solid solutions, which may be metastable.  Significant discrepancies are found in the literature between the (x,y)-compositional phase diagrams for LMFP determined computationally versus experimentally (by chemical delithiation of large particle powders).^8,9

We demonstrate here, by systematic screening of the electrochemical driven phase transitions in a series of LiMn_yFe_1-yPO₄ (y = 0.1, 0.2, 0.4, 0.6 and 0.8) powders, that this material exhibits a completely different phase transformation mode dominated by formation of metastable solid solutions for nanoscale LMFP compared to the binary lithiation states within the extremely well-studied case of LFP. The screening is conducted by operando synchrotron radiation powder X-ray diffraction (SR-PXD) during charge and discharge at different C-rates on powders with particle size of ~50, 100 and 150 nm. Through Rietveld refinement (>1000 PXD patterns have been refined for this study) the misfit strains during phase transformations are examined, revealing small elastic misfits between phases within the extended solid solution regime compared to either pure LiFePO₄ or LiMnPO₄. Hence, on the basis of the time- and state-of-charge dependence of the olivine structure parameters, we propose a coherent transformation mechanism. Finally, we bring evidence that the observed metastability is enabled by particle size reduction to the nanoscale. 

This work was supported by DOE project number DE-SC0002626. Work at APS was supported by DOE Contract No.DE-AC02-06CH11357. D.B.R. acknowledges the Carlsberg Foundation for funding.

(1)Meethong et al. Electrochem. Solid-State Lett. 2007,10,A134. (2)Meethong et al. Adv. Funct. Mater. 2007,17,1115. (3)Meethong et al. Chem. Mater. 2008,20,6189. (4)Meethong et al. Adv. Funct. Mater. 2009,19,1060 (5)Chueh et al. Nano Lett. 2013,13,866. (6)Chung et al. Nat. Mater. 2002,1,123. (7)Y.-H. Kao, Ph.D. thesis, MIT, 2011. (8)Yamada et al. J. Electrochem. Soc. 2001,148,A1153. (9)Malik et al. Phys. Rev. B 2009,79,214201.