5.3 Stress–strain and ideal strength
These are unexecuted teaching inputs and starting settings to test. Original figures are schematics, not computed results. Use licensed VASP and PAW data, replace every placeholder, record the executable version and validate convergence.
5.3.1 Model, units and provenance
Use eV for energy, Å for length and eV/Å for force; 1 kbar = 0.1 GPa. State normalization per atom, molecule, primitive cell or simulation cell. Record PAW identifiers, release, ZVAL, ENMAX and permitted hashes; never redistribute POTCAR. SCF convergence addresses the chosen electronic problem; convergence of the target property requires separate tests.
Original schematic. Curves explain concepts; blank data areas await verified learner results. No calculation is claimed.
5.3.2 Worked case: procedure, interpretation and checks
Physical target
Elastic constants describe the initial slope near zero strain. Ideal strength asks how much stress a perfect crystal can sustain along a specified loading path before it loses stability. It is a theoretical property of a defect-free, constrained, usually zero-temperature model. It is not a direct prediction of a tensile-test yield stress, which can be controlled by dislocations, cracks, surfaces, grain boundaries, temperature, and strain rate.
The loading protocol is part of the answer. Uniaxial strain fixes lateral dimensions. Uniaxial stress lets transverse dimensions relax until transverse stresses vanish. They need not give the same stress peak. Define engineering strain e=(L−L₀)/L₀; logarithmic strain is ln(L/L₀). At large strain specify the stress measure as well, rather than mixing engineering stress with the Cauchy stress returned by a periodic-cell calculation.
Workflow
Choose a loading direction and rotate or construct the cell so that the intended crystallographic direction is represented transparently. Relax the zero-load reference. Apply incremental deformation gradients to the reference cell; store both cumulative strain and increment. At each strain, relax permitted internal coordinates. For an explicitly clamped lateral cell, ISIF=2 is appropriate. For zero transverse stress, use a validated constrained-cell optimizer or an outer loop over transverse lattice components; unconstrained ISIF=3 can release the applied axial strain and invalidate the experiment.
Start with moderate steps, then refine near curvature changes or a stress maximum. Keep the previous relaxed geometry as one initial guess, but also test symmetry-broken perturbations. A primitive cell with enforced symmetry may suppress the mode that would actually destabilize a larger crystal. Run more than one allowed structural branch if the energy landscape admits competing pathways. Record all six stress components, energy, volume, atomic displacements, magnetization where relevant, and optimizer convergence.
Input delta
This block neither creates strain nor imposes zero transverse stress. Those are geometry-generation and constraint tasks outside these tags. Use a tight static run to read stress at the converged geometry. For a simple axial deformation with fixed transverse area A₀, engineering axial stress and Cauchy stress coincide only under the appropriate area relation; in general σeng = σtrue(A/A₀). For full finite-strain work, use a consistent deformation-gradient/stress pair and publish its definition.
Interpretation and limits
A peak in stress is a candidate ideal strength on the sampled branch. It does not prove that this branch was stable all the way to the peak. A shear or phonon instability may occur earlier. Inspect elastic response or phonons at selected strained structures, allow larger-cell modes, and visualize bond rearrangements. A discontinuity can be a real phase transformation or a failed relaxation; distinguish them by restarts, smaller strain steps, and independent initial guesses.
Check stress convergence at a strain near the peak, not only at equilibrium. Compare loading and unloading to identify metastability, but do not interpret static hysteresis as experimental rate-dependent dissipation. Treat finite-temperature strength as a separate free-energy/dynamical problem.
Exercise: compute the same direction under lateral clamping and transverse-stress relaxation. Predict how Poisson contraction changes the energy and stress, then test the prediction. Report protocol sensitivity before comparing with experiment.
5.3.3 Unexecuted inputs and analysis scaffolds
These are unexecuted teaching inputs and starting settings to test. Original figures are schematics, not computed results. Use licensed VASP and PAW data, replace every placeholder, record the executable version and validate convergence.
5.3.3.1 Input block 1
# Fixed prescribed deformation; only atoms relax
IBRION = 2
ISIF = 2
NSW = 200
EDIFFG = -0.003
ISYM = 0
# Cell deformation and any transverse-cell search are external workflow steps.
5.3.4 Related learning paths
- 1.1 Four input files, one physical question
- 1.2 A convergence laboratory with an error budget
- 5.2 Elastic tensor from finite strains
- 6.1 Finite-displacement phonons with Phonopy