Herein we describe a technique for preparing samples from cycled Li-ion batteries with minimal contact with atmospheric water for examination by TEM and to provide the results of an analysis of the SEI in Li-ion cells. The electrode samples were prepared in a glove box by manipulation with a diamond tipped pen to carefully dislodge particles directly onto the TEM sample holder. Electrodes were extracted from Li-ion cells that contained a cathode of high capacity, manganese rich NCM (HCMR-NCM). Nanometer-sized MnF2 crystal particles embed themselves in the SEI layer of the anodes as observed and confirmed by HRTEM lattice image analysis and EELS. Cross-sections of the SEI layer reveal that the thickness of the SEI and propensity for MnF2 crystal deposition is non-uniform.

LiNi0.5Mn1.5O4 cathode material, which has a higher working voltage (4.7 v) than NCM and a moderate specific capacity (148 mAh/g theoretical), has been studied to understand the source of capacity fade during the first 100 cycles in a half cell. The work mainly consisted of high resolution TEM observations and analysis of the surface microstructural properties, before and after cycling. We found that the pristine material consisted almost entirely of large FCC spinal domains but with cycling appears small simple cubic spinel domains at the surface. It is proposed that these small changes of the surface microstructure leads to impedance rise that results in the premature arrival to the upper cutoff voltage of 4.85V during charging and the subsequent loss of capacity with cycling.

Heterostructures and interfacial defects in a 40-nm-thick SrTiO3 (STO) film grown epitaxially on a single-crystal MgO (001) were investigated using aberration-corrected scanning transmission electron microscopy and geometric phase analysis. The interface of STO/MgO was found to be of the typical domain-matching epitaxy with a misfit dislocation network having a Burgers vector of ½ aSTO ⟨100⟩. Our studies also revealed that the misfit dislocation cores at the heterogeneous interface display various local cation arrangements in terms of the combination of the extra-half inserting plane and the initial film plane. The type of the inserting plane, either the SrO or the TiO2 plane, alters with actual interfacial conditions. Contrary to previous theoretical calculations, the starting film planes were found to be dominated by the SrO layer, i.e., a SrO/MgO interface. In certain regions, the starting film planes change to the TiO2/MgO interface because of atomic steps at the MgO substrate surface. In particular, four basic misfit dislocation core configurations of the STO/MgO system have been identified and discussed in relation to the substrate surface terraces and possible interdiffusion. The interface structure of the system in reverse—MgO/STO—is also studied and presented for comparison.

The TEAM 0.5 electron microscope is employed to demonstrate atomic resolution phase contrast imaging and focal series reconstruction with acceleration voltages between 20 and 300 kV and a variable dose rate. A monochromator with an energy spread of ≤0.1 eV is used for dose variation by a factor of 1,000 and to provide a beam-limiting aperture. The sub-Ångstrøm performance of the instrument remains uncompromised. Using samples obtained from silicon wafers by chemical etching, the [200] atom dumbbell distance of 1.36 Å can be resolved in single images and reconstructed exit wave functions at 300, 80, and 50 kV. At 20 kV, atomic resolution <2 Å is readily available but limited by residual lens aberrations at large scattering angles. Exit wave functions reconstructed from images recorded under low dose rate conditions show sharper atom peaks as compared to high dose rate. The observed dose rate dependence of the signal is explained by a reduction of beam-induced atom displacements. If a combined sample and instrument instability is considered, the experimental image contrast can be matched quantitatively to simulations. The described development allows for atomic resolution transmission electron microscopy of interfaces between soft and hard materials over a wide range of voltages and electron doses.

This paper reports on the defect structures formed upon strain relaxation in pulsed laser-deposited complex oxide superlattices consisting of the ferromagnetic metal, La0.67Sr0.33MnO3, and the antiferromagnetic insulator, La0.67Sr0.33FeO3. Atomic resolution scanning transmission electron microscopy and electron energy loss spectroscopy were used to characterize the structure and chemistry of the defects. For thinner superlattices, strain relaxation occurs through the formation of 2-D stacking faults, whereas for thicker superlattices, the prolonged thermal exposure during film growth leads to the formation of nanoflowers and cracks/pinholes to reduce the overall strain energy.

FePt nanoparticles are promising materials for high-density magnetic data storage media [1] and bio-medical applications such as drug-targeting and hyperthermia [2]. To understand their magnetic properties [3] it is essential to get insights into the lattice structure of isolated nanoparticles which influence the magnetic behavior.

Typically, lattice fringes are observed with high-resolution transmission electron microscopy (HR-TEM). In this case delocalization effects disturb imaging of the lattice structure in particular if 2 to 6 nm small nanoparticles are involved. Therefore, FePt nanocrystals were investigated by reconstructing amplitude and phase of the scattered electron wave from a focal series of HRTEM images, which can produce delocalization free and direct images of the crystal structure [4]. The formation of 5-fold twinned structures of 3 to 7 nm face-centered cubic FePt nanocrystals is investigated that were grown from a colloidal solution [1]. The results are compared with abinitio density functional (DFT) calculations of FePt particles with a diameter of larger than 2 nm. Image simulations were performed with the Accelrys Cerius2 software package (Version 4.6). Good agreement between the ab-initio calculations and the experimental data is found.

Nanowires or porous films grown on a substrate normally lack mechanical strength, and may be subject to damage during specimen preparation. When we made cross-sectional TEM specimen for this type of sample, we modified the traditional method by covering the sample with epoxy to improve the film strength, and applying single-section ion milling to protect the film from over-milling.

The sample surface is first covered with G1 epoxy. We choose G1 for this application because it is relatively thick and cures at low temperature. For samples with a dense-growth of nanowires or a thick porous film, a brief moment in vacuum helps to get rid of the air bubbles in the epoxy. The glue is cured at 100 degree C for 10 minutes, until its color turns to a reddish brown. To remove the excess glue and flatten the surface, the sample is then ground and polished until the glue is less than 0.1 mm thick.