Find your next learning step
Returning the external strain to zero does not guarantee that every local neighbourhood returns to its starting geometry. Cyclic loading separates reversible non-affine motion, delayed recovery, training and irreversible rearrangement through cycle-by-cycle return maps.
Open the chapter contents · Polymer Science root · Gold Standard of Education · How X Works
Chapter contents
Choose the route closest to the question, or read from Chapter 1 for the complete argument. The chapters remain open on the page; these grouped routes are an index.
Route 1 · Define a cycle and reference
Route 2 · Work the return map
Route 3 · Separate memory from damage
Route 4 · Check evidence and limits
Route 5 · Ask and continue
CHAPTER 1 OF 12 · DEFINE A CYCLE AND REFERENCE
1. A cycle needs a declared reference state
Define cycle n by its loading path, maximum strain, unloading path, hold times, rate and temperature. Let state Rn be the measured configuration after returning the imposed strain to the chosen reference value. Comparing peaks alone cannot reveal whether the material returned locally.
The guide on neighbour turnover and irreversibility establishes correspondence. Here, those identities are followed from R0 through repeated return states.
CHAPTER 2 OF 12 · DEFINE A CYCLE AND REFERENCE
2. Residual non-affinity is a return-map quantity
Compute D²min between R0 and Rn using a fixed, reported neighbourhood rule. A nonzero residual after macroscopic strain returns to zero indicates local mismatch relative to the fitted affine return. It may arise from irreversible rearrangement, delayed viscoelastic recovery, drift or tracking error.
Also calculate cycle-to-cycle D²min between Rn−1 and Rn. A large first-cycle change followed by smaller changes describes training differently from steady accumulation.
| Comparison | Question |
|---|---|
| peak to peak | does the loaded configuration repeat? |
| R0 to Rn | how far has the return state drifted? |
| Rn−1 to Rn | is the next cycle still changing the state? |
CHAPTER 3 OF 12 · WORK THE RETURN MAP
3. Work a three-cycle return ledger
Consider an invented tracked neighbourhood with normalized residual D²min values from R0 of 0.00, 0.18, 0.23 and 0.24 after cycles 0–3. Cycle-to-cycle residuals are 0.18, 0.06 and 0.02. The cumulative residual grows, while incremental change falls.
This pattern is consistent with rapid first-cycle reorganization followed by approach to a repeatable trained state. It is not proof of a microscopic mechanism. Stress, neighbour recovery, energy and uncertainty must agree.
| Return state | D²min from R0 | D²min from previous return |
|---|---|---|
| R0 | 0.00 | — |
| R1 | 0.18 | 0.18 |
| R2 | 0.23 | 0.06 |
| R3 | 0.24 | 0.02 |
CHAPTER 4 OF 12 · WORK THE RETURN MAP
4. Hysteresis and residual geometry answer different questions
Stress–strain loop area measures mechanical energy dissipated per cycle under the stated convention. Residual D²min measures local geometric mismatch. A loop can narrow while a changed cage persists; a viscoelastic loop can also enclose area without permanent neighbour turnover.
Report both rather than translating one into the other. Correct force, displacement and phase offsets before integrating loop area.
CHAPTER 5 OF 12 · WORK THE RETURN MAP
5. Changed context: add a recovery hold
Repeat the return measurement after a longer zero-load hold. If D²min from R0 falls from 0.24 to 0.12 without reversing the loading path, part of the earlier residual was delayed recovery on that timescale.
If it stays near 0.24 while controls show stable registration, the evidence for a persistent structural change strengthens. “Permanent” still requires a stated observation window; absence of recovery in ten minutes is not absence of recovery forever.
CHAPTER 6 OF 12 · SEPARATE MEMORY FROM DAMAGE
6. Cage memory can be directional
A trained material may respond differently when reloaded in the trained direction, the reverse direction or an orthogonal direction. Compare identical amplitudes and rates while changing only direction. Directional response can reveal anisotropy left by the loading history.
Do not call every directional difference memory. Sample alignment, chain orientation, residual stress and boundary conditions are physical alternatives that need measurement.
CHAPTER 7 OF 12 · SEPARATE MEMORY FROM DAMAGE
7. Cyclic loading can anneal or rejuvenate
Below some regimes, repeated loading may move a glass toward lower-energy or more repeatable states; larger amplitudes can produce irreversible rearrangement, yielding or damage. Polymer-glass experiments also distinguish overaging and rejuvenation through segmental dynamics.
These outcomes depend on amplitude, temperature, rate, age and material architecture. A single “cycling improves the material” rule would be false.
CHAPTER 8 OF 12 · SEPARATE MEMORY FROM DAMAGE
8. Tracking and reference choices remain active
Apply the same identity audit on every cycle. Selective track loss can make later return states appear quieter. A refreshed neighbour set answers whether the current cage is affine, not whether the original cage returned.
Archive coordinates and transformations so R0-to-Rn calculations can be recomputed. Do not chain only adjacent displacements; small registration errors can accumulate.
CHAPTER 9 OF 12 · CHECK EVIDENCE AND LIMITS
9. Limits and misconceptions
Residual non-affinity is not synonymous with damage. Reversible heterogeneity, delayed recovery and noise contribute. Conversely, a small mean can hide a localized shear band or craze.
The worked values are dimensionless teaching numbers. D²min scaling, cutoff, Δt, temperature and localization precision must accompany any real value.
CHAPTER 10 OF 12 · CHECK EVIDENCE AND LIMITS
10. A practical evidence check
Use a no-load control, repeat specimens and at least one recovery hold. Save stress–strain loops, energy or density proxies, neighbour retention and spatial D²min maps at every return state.
Plot cumulative and incremental residuals against cycle number. Test amplitude and direction without changing rate or temperature. Inspect whether high-residual regions recur or migrate.
CHAPTER 11 OF 12 · ASK AND CONTINUE
11. Questions readers should ask
Does returning strain to zero mean the structure returned?
No. The boundary condition returned; local geometry and stress may not have.
Does a narrowing loop prove healing?
No. It can reflect training or viscoelastic change while residual structure remains.
When is a residual irreversible?
When it persists beyond stated recovery and uncertainty checks and is supported by return geometry, neighbour and mechanical evidence.
CHAPTER 12 OF 12 · ASK AND CONTINUE
12. Continue through verified routes
Begin with the Polymer Science root, Gold Standard of Education and How X Works. Connect Non-Affine Relaxation, D-Squared-Min Scale Dependence and the neighbour-turnover guide above.
For specialist evidence, read the primary study on cyclic loading and segmental dynamics in PMMA glass and molecular simulations of crazes under cyclic loading. The return ledger is an original teaching example.
Top · Contents · Previous · How X Works