Search

Atom probe tomography (APT) can yield three-dimensional tomographic images at atomic-scale resolution and low-AMU elements such as Li are readily observed, making it a powerful tool for exploring battery materials interfaces. However, it is difficult to prepare APT specimen tips containing the interface of interest starting with typical particle-based battery electrodes. Here we demonstrate a methodology for reliable APT imaging of battery interfaces in which a thin film electrode geometry is used to provide well-controlled planar interfaces that are ideal for APT sample preparation and imaging. LiFePO4(LFP) thin film electrodes, synthesized using pulsed laser deposition (PLD), were studied as an example system, with standard Li-salt electrolytes. For the results to be applicable to conventional particulate electrodes, it is important to obtain representative thin film structure and electrochemical characteristics. Thus, the effects of PLD conditions including substrate temperature, substrate crystallinity, target composition, and deposition time (number of laser pulses) on the thin film’s crystallographic texture, morphology, and electrochemical performance were studied. Optimized LFP film showed good crystallinity with low-C-rate capacity of ∼90 mAh g−1. Initial APT three-dimensional imaging of the LFP/electrolyte interface shows an ∼10 nm cathode-electrolyte interphase layer that is enriched in F and Li.

Reversible cycling of MnO2cathodes remains a challenge in alkaline rechargeable battery development. In prior work, we have developed new platelet shaped MnO2nanoparticles with stable performance in LiOH electrolyte for over 50 cycles. In this study, the effects of sub-nanoparticle organization of MnO2polymorphs on cycling performance, phase activation, and charge/discharge mechanisms in LiOH electrolyte are investigated by ex situX-ray powder diffraction (XRD). Different phase compositions with the same particle morphology are achieved by annealing as-synthesized nanoplatelets composed of 50% ramsdellite (R-MnO2) and 50% akhtenskite (ε-MnO2) polymorphs. Thermal treatments result in a gradual change of polymorph composition, due to reorganization of the 1 × 2 channels in ramsdellite and akhtenskite phases to 1 × 1 channels with initial appearance of the gamma phase (γ-MnO2), followed by complete conversion of ramsdellite to pyrolusite (β-MnO2) at 400 °C. Electrochemical activity of thermally treated nanoparticles is correlated to the phase compositions before and after cycling. Rietveld refinement of the XRD patterns suggests material activation in the initial cycles through intercalation of Li+and H+ions into 1 × 2 channels, resulting in lattice expansion in both akhtenskite and ramsdellite, while intercalation of ions into structural faults present in akhtenskite and gamma phases results in amorphization of the active material. The detailed mechanism of the polymorph conversion during annealing as well as phase activation and reversible redox activity are discussed. For the first time, it is clearly demonstrated that the electrochemical activity of MnO2material strongly depends not only on the lattice structure of individual polymorphs but also on the sub-nanoparticle architecture, including surrounding polymorphs and interphases.

In this work, a comprehensive equivalent circuit model (ECM) was developed for the a commercial 26650 LiFePO4/Li4Ti5O12cell in which the main processes at the anode and cathode electrodes and their contributions to the full-cell impedance were investigated and quantified separately. To unambiguously separate the anode and cathode processes, EIS spectra from Li4Ti5O12/Li and LiFePO4/Li half cells were modeled at different state of charge (SOC) and SOC history and the fit results were used to develop the full-cell ECM. The distribution of relaxation times (DRT) analysis was used in all cases to better resolve the frequencies of the processes, and thereby help their accurate identification. The LiFePO4/Li4Ti5O12-cell impedance was found to be significantly dependant on the SOC history, i.e. it changes slightly only at low SOCs during charge, whereas, it changes significantly during discharge. The SOC-dependence of the full-cell impedance was found to be mainly due to the decrease of the LiFePO4electrode's charge transfer resistance with SOC increase. While Li4Ti5O12electrode processes did not show a significant change with SOC history, it was revealed that the SOC-histoty dependence of the LiFePO4/Li4Ti5O12full cell was also mainly due to the variation of charge transfer frequency and resistance of LiFePO4electrode.