Both bars are the same crystal, only the length scale differs. The left bar is millimeter-sized and, like any real bulk crystal, riddled with dislocations — line defects that glide under modest shear stress and let the lattice slip layer over layer. That glide is macroscopic plastic deformation: the bar yields early and keeps stretching without needing more stress.
The right bar is a nanowire a few hundred atoms across. A volume that small usually contains few or no dislocations to begin with, and any that were there tend to glide straight out through the free surface during growth or the first instants of loading — "dislocation starvation." With no mobile defects left to carry plastic flow, the nanowire has no choice but to stretch elastically, storing stress right up toward the theoretical strength of a perfect lattice, until it snaps.
macro: σ < σ_yield → strain = σ/E (elastic)
σ ≥ σ_yield → strain = σ/E + k·(σ−σ_yield) (dislocation glide)
nano: σ < σ_fracture → strain = σ/E (elastic only)
σ ≥ σ_fracture → sudden brittle fracture, σ → 0
- Applied load — ramps engineering stress on both samples simultaneously, from 0 up to 5 GPa.
- Auto-ramp — drives the load up automatically so you can watch both curves develop in real time.
- Macro state — Elastic → Yielding (dislocations gliding, bar visibly elongates and glows).
- Nano state — Elastic → Fractured (sudden brittle snap, no warning plastic stage).
Real-world relevance: this size effect — "smaller is stronger" — is why nanowires, nanopillars and thin films used in MEMS/NEMS devices and nanoelectronics can approach the theoretical strength of a perfect crystal, while also failing catastrophically and without warning once that limit is reached, which is exactly why nanoscale reliability testing (in-situ TEM/SEM nanocompression, fatigue cycling) matters so much for device design.