Low Voltage (10 to 30 keV) CRYO-STEM-EELS: Another Step Toward a Damage-free Mapping of Li in Beam Sensitive Materials

Electron beam damage in electron microscopes is becoming more and more problematic in material research with the increasing demand of characterization of new beam sensitive material such as Li based compound used in batteries. Mainly two types of damage can be induced by the electron beam: radiolysis and knock-on damage. To avoid radiolysis damage caused by the inelastic interaction between a beam electron and a valence electron from one of the material’s atoms, the use of cryo-microscopy is becoming the conventional method for battery material characterization. By cooling the sample to cryogenic temperatures, radiolysis damage is reduced, however, it generally uses a high-energy electron beam (200-300 keV) which can be problematic for low Z materials. Through elastic interactions, a high-energy electron from the beam can transfer its energy to an atom’s nucleus and cause an atomic displacement which is called knock-on damage [2]. To prevent this type of damage, it is well known that the beam energy needs to be below the energy threshold of atomic displacement [3, 4]. This means that to reduce the beam damage on beam sensitive material, we not only need to use cryo-stage holder, but also turn to low beam voltage electron microscopy. Using the Hitachi SU-9000EA microscope, which is a low-voltage cold-field emission STEM-EELS instrument with a 0.5eV energy resolution [5], EELS spectra of TiN were acquired at 30, 20 and 10 keV. Fig. 1a clearly shows both N K and Ti L2,3 edge for all three beam voltages. Working at 10 keV in transmission is at the limit of electron transparency which explains the poor signal-to-noise ratio. The validity of the EELS quantification method with a 30 keV accelerating voltage was investigated. Since the N K edge (402 eV) overlap the Ti L3,2 edge (456 eV), the model-based approach [6] for quantification was preferred to Egerton’s integration method [7]. As shown in Fig. 1b, the quantification is accurately achieved within the ∼10% error coming from the Hartree-slater partial cross-sections of the nitrogen K edge [8]. Following the promising results of the TiN, two Li-based compounds, lithium titanate (LTO) and spodumene (LiAl(SiO2)2), were investigated using the 30 keV electron beam under cryogenic conditions (-160'C). Fig. 2a shows the well-defined background subtracted Li K, Ti L2,3 and the O K edge of LTO and no apparent damage was observed after the acquisition. The same procedure was done on a thin film of spodumene produced by FIB and the Li K, Al L2,3 and Si L2,3 edges were easily seen on the spectra shown in Fig. 2b. Using non-negative matrix factorization (NMF), it has been possible to isolate the Li signal into one component of the factorization (Fig. 3a). From the NMF, an intensity map related to the Li signal component was generated and can be related to the distribution of Li within the mineral. An unexpected vein-like distribution of Li can be seen in Fig. 3b which could be the sign of the diffusion of the Li atoms caused by the FIB in the sample preparation procedure. a) N k edge and Ti L2,3 edge of TiN acquired with a 10keV (blue), 20keV (orange) and 30 keV (green) electron beam. b) EELS quantification results of TiN using a 30 keV electrons beam a) Li K, Ti L2.3 and O K edge of lithium titanate acquired with a 30 keV electron beam. b) Li K, Al L2.3 and Si L2,3 edge of spodumene acquired with a 30 keV electron beam. a) Lithium signal isolated from the spodumene spectrum by NMF. b) Loading intensity map of the lithium signal obtained by NMF showing the Li distribution in spodumene.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.