Electrochemical strain microscopy (ESM) has been developed with the aim of measuring Vegard strains in mixed ionic-electronic conductors (MIECs), such as electrode materials for Li-ion batteries, caused by local changes in the chemical composition. In this technique, a voltage-biased AFM tip is used in contact resonance mode. However, extracting quantitative strain information from ESM experiments is highly challenging due to the complexity of the signal generation process. In particular, electrostatic interactions between tip and sample contribute significantly to the measured ESM signals, and the separation of Vegard strain-induced signal contributions from electrostatically induced signal contributions is by no means a trivial task. Recently, we have published a compensation method for eliminating frequency-independent electrostatic contributions in ESM measurements. Here, we demonstrate the potential of this method for detecting Vegard strain in MIECs by choosing Cu$$_2$$ 2 Mo$$_6$$ 6 S$$_8$$ 8 as a model-type MIEC with an exceptionally high Cu chemical diffusion coefficient. Even for this material, Vegard strains are only measurable around and above room-temperature and with proper elimination of electrostatics. The analyis of the measured Vegards strains gives strong indication that due to a high charge transfer resistance at the tip/interface, the local Cu concentration variations are much smaller than predicted by the local Nernst equation. This suggests that charge transfer resistances have to be analyzed in more detail in future ESM studies.

In piezoresponse force microscopy and electrochemical strain microscopy (PFM, ESM), not only nanoscale electromechanical surface displacements (e.g., Vegard strain in case of ESM) are amplified in contact resonance; global cantilever capacitive forces are as well. In addition, other nanoscale nonelectrical contact mechanics could contribute to the contrast formation, too. Here we propose a method to separate these contributions by using the band excitation method together with an amplitude modulated high-frequency electric potential applied to the cantilever. Compared to the conventional DC biased low-frequency AC contact resonance mode, this allows us to determine voltage and frequency-dependent nanoscale surface responses quantitatively, because the capacitive components are deducted. Numerical simulations based on the Euler−Bernoulli equation together with experiments on Li-ion conducting glass ceramics (LICGCs) and on the mixed Cu-ion/electron-conducting material Cu 2 Mo 6 S 8 demonstrate the advantages of this approach.