From surface hardening of steels to doping of semiconductors, atom insertion in solids plays an important role in modifying chemical, physical, and electronic properties of materials for a variety of applications. High densities of atomic insertion in a solid can result in dramatic structural transformations and associated changes in mechanical behavior: This is particularly evident during electrochemical cycling of novel battery electrodes, such as alloying anodes, conversion oxides, and sulfur and oxygen cathodes. Silicon, which undergoes 400% volume expansion when alloying with lithium, is an extreme case and represents an excellent model system for study. Here, we show that fracture locations are highly anisotropic for lithiation of crystalline Si nanopillars and that fracture is strongly correlated with previously discovered anisotropic expansion. Contrary to earlier theoretical models based on diffusion-induced stresses where fracture is predicted to occur in the core of the pillars during lithiation, the observed cracks are present only in the amorphous lithiated shell. We also show that the critical fracture size is between about 240 and 360 nm and that it depends on the electrochemical reaction rate.
A model of micro-pillar plasticity controlled by the nucleation of dislocations at free surfaces was developed using methods of dislocation dynamics innovated by Johnston and Gilman 50 years ago. It is shown that the stress dependence of the rate of nucleation can be related to the dependence of the flow strength on the diameter of the pillar. The steady state flow stress in this model depends only on the kinetics of nucleation, whereas the transient deformation behavior depends on the initial dislocation density and the dislocation mobility, as well as the rate of nucleation. The model can describe not only the dependence of the strength on pillar diameter, but also recent experiments on fcc gold crystals, wherein pre-straining leads to softening and annealing leads to hardening. The model does not account for the stochastic, jerky nature of plastic flow in micro-pillars and is not meant to apply to pillars large enough to support substructure formation.
Microscale electrodes are rapidly becoming critical tools for neuroscience and brain-machine interfaces (BMIs) for their high spatial and temporal resolution. However, the mechanics of how devices on this scale insert into brain tissue is unknown, making it difficult to balance between larger probes with higher stiffness, or smaller probes with lower damage. Measurements have been experimentally challenging due to the large deformations, rapid events, and small forces involved. Here we modified a nanoindentation force measurement system to provide the first ultra-high resolution force, distance, and temporal recordings of brain penetration as a function of microwire diameter (7.5 µm to 100 µm) and tip geometry (flat, angled, and electrosharpened). Surprisingly, both penetration force and tissue compression scaled linearly with wire diameter, rather than cross-sectional area. Linear brain compression with wire diameter strongly suggest smaller probes will cause less tissue damage upon insertion, though unexpectedly no statistical difference was observed between angled and flat tipped probes. These first of their kind measurements provide a mechanical framework for designing effective microprobe geometries while limiting mechanical damage.