Point and grain inspector
Hover over the map to inspect a point. Click to lock a grain selection.
Skip to workspace
Browser analysis
Local scientific analysis
Reference frameNot loaded Moving frameNot loaded
Moving pointsClick source landmark Reference pointsClick matching target landmark
Local scientific workflow
Metric-native engine
Not evaluated
Not trained
Checking engine
Checking engine
Checking engine
Checking engine
← Metallurgy workspaces
Microstructure & Crystal Plasticity
Research-grade workflow studio.
Connect microscopy, transformation crystallography, parent reconstruction, microstructure evolution, process calibration and native crystal-plasticity solvers in one controlled interface.
Checking compute enginesBrowser and server capabilities are verified before a run.
EBSD · texture · mechanics · reconstruction · RVE
Microstructure Studio 3.0
Analyze indexed orientation maps with explicit provenance, quantify grains and texture, screen slip activity, and create periodic solver-ready voxel RVEs without sending data to a third party.
Orientation and property map
Hover over a point or select a grain for crystallographic and mechanics details.
Load an EBSD mapPhase-aware cleaning, grains, KAM/GROD/GOS, boundaries, mechanics and texture are calculated locally.
Scientific provenance
Every cleaning, symmetry and reconstruction choice is included in exports.
No processing record yet.
{001} pole figure
Equal-area, antipodal, MRD-normalized KDE.
{011} pole figure
Equal-area, antipodal, MRD-normalized KDE.
{111} pole figure
Equal-area, antipodal, MRD-normalized KDE.
Inverse pole figure
Specimen Z in the crystal fundamental region.
ODF section φ₂ = 0°
ODF section φ₂ = 45°
ODF section φ₂ = 65°
Texture components
Fractions and deterministic bootstrap confidence intervals.
| Component | Fraction | 95% interval |
|---|---|---|
| Run texture analysis. | ||
Custom pole family
Enter a crystal direction family for the selected symmetry.
Crystal lattice calculator
Convert direct and reciprocal indices with a general triclinic metric. Four-index hexagonal directions and planes are accepted.
Enter lattice parameters and indices.
Interpretation safeguards
- Plane normals use the reciprocal lattice; directions use the direct lattice.
- d-spacing is reported in the same length unit as a, b and c.
- A slip direction should lie in its plane, giving a 90° angle to the plane normal.
- Reference-frame and index-convention choices remain part of the exported provenance.
Boundary misorientation
No distribution available.
Grain-size distribution
No distribution available.
Grain orientation spread
No distribution available.
Aspect ratio
No distribution available.
Boundary and junction network
No boundary network available.
GOS versus grain size
Advanced topology
Run reconstruction.
Spatial correlations
Run reconstruction.
General stress tensor and fibre screening
Maximum Schmid-factor distribution
Run mechanics analysis.
KAM-derived GND proxy
Run mechanics analysis.
Interpretation: Schmid factors are geometric uniaxial factors. The GND value is a KAM-based screening proxy, not an HR-EBSD dislocation-density inversion.
Scientific boundary: MatAMet performs orientation-space post-processing and reproducible screening. It does not re-index diffraction patterns, perform HR-EBSD elastic-strain inversion, or claim exact equivalence to MTEX spherical-harmonic analysis.
Voxel slice
Not generated
Build a reproducible RVEWeighted, periodic voxel grains or an EBSD-derived extrusion can be packaged for the native engines.
RVE validation
Build an RVE.
Directional chord lengths
Build an RVE.
Target versus achieved volumes
Build an RVE.
Phase two-point correlation
Build an RVE.
Solver-ready means structurally prepared, not calibrated: constitutive parameters, units, boundary conditions and exact version-specific keywords must be reviewed before a native run.
SEM · BSE · EBSD · EDS · phase maps · uncertainty · physics
SynchroMet Correlative Studio 4.0
Register multimodal microscopy, segment phases, quantify particles and interfaces, propagate uncertainty, and hand measured microstructures directly to MatAMet reconstruction and crystal-plasticity workflows.
Open interoperability:scikit-imageOpenCVilastikorixTrueEBSDAlgorithms here are independently implemented and exports preserve provenance.
Load the target imageThe reference frame should have the highest useful spatial resolution.
Load a correlated modalityUse recognizable particles, scratches, pores or boundary junctions as landmarks.
Project auditWaiting for data.
Control-point modeChoose a point in the moving frame first.
Particle-size distribution
Nucleation-site association
Boundary-distance distribution
Spatial correlation L(r)
Object-level results
| ID | ECD | Aspect | Feret max | Circularity | Convexity | Solidity | Nearest edge | Associated sites |
|---|
Particle-resolved Zener contribution
PSN screening
Boundary/site-specific pinning
Run the physics calculation.
Interpretation ledger
Run the physics calculation after particle correlation.
Measured change
| Metric | State A | State B | Δ | Change |
|---|
Reproducible exports
Transformation history · recrystallization · precipitation · processing
MatAMet M³ Decision Twin
Reconstruct mechanisms, quantify uncertainty, compare competing hypotheses, identify robust process windows and hand evidence-graded states to the existing native solvers.
Original implementation — no copied ARPGE, MTEX or proprietary source.Traceable models — equations, conventions and calibration state remain visible.Native escalation — full-field jobs remain with VPSC, DAMASK, Fierro, MOOSE/CPFEM or phase-field solvers.
Metric tensor · cross tensor · direct crystal-basis rotations
Crystallographic Quaternion Laboratory
Compose rotations, rotate crystallographic vectors and inspect EBSD misorientation axes directly in non-Cartesian crystal bases without treating an arbitrary Cartesian crystal frame as fundamental.
Basis-native: scalar products use M and cross products use X = √det(M) M⁻¹.Cross-validated: every result is compared with an independent structure-tensor/Cartesian route.Convention-aware: active and passive interpretations remain explicit.
Metric, cross and structure tensors
The metric and cross tensors are convention-independent. The structure tensor shown here is only the conventional Cartesian validation frame.
Quaternion product
Run a composition.
Rotated crystallographic vector
Run a composition.
Independent validation
The direct metric/cross-tensor route will be compared against a conventional Cartesian structure-tensor route.
Spherical metric verification
Verify the spherical cosine and sine laws for three directions using only the crystal metric.
Choose three non-coplanar directions.
Reconstructed parent map
Parent IDs are spatially split by child-grain connectivity. Unassigned grains remain neutral.
Parent population
Residual distribution
Parent grains and evidence
| Parent | Children | Area | Mean residual | Max residual | Confidence |
|---|
Decision-grade result
Orientation-relationship competition & reconstruction authority
Area-weighted coverage, residual tails, ambiguity, variant richness and spatial coherence are combined without treating the score as proof of a unique mechanism.
—Run reconstruction
Competing hypotheses
| Relationship | Evidence | Area coverage | P95 residual | Confidence | Topology | Relative weight |
|---|
Required authority checks
Run a reconstruction first.
Continuous distortion path
Principal stretches
Transformation tensors
Variant compatibility
Calculate the transformation first.
Mechanism evidence
Compatibility robustness & interpretation boundary
Combines λ₂ compatibility, rank-one quality, volume change and lattice-parameter sensitivity. It never upgrades an approximate plane into a PTMC solution.
—Calculate transformation
Lattice uncertainty envelope
Interpretation and next validation
Calculate a transformation first.
Recrystallized fraction
Stored energy and pressure
Evolution table
| Time | T | ρ | Pressure | Nuclei density | Xrex | Mean size |
|---|
Predictive envelope
Uncertainty, nucleation-site contributions & kinetic bottleneck
Runs a deterministic parameter ensemble around the active model and reports t10/t50/t90, final uncertainty, site contributions and dominant sensitivity drivers.
—Run ensemble
5–50–95% recrystallization envelope
Nucleation-site contribution
Dominant uncertainty drivers
| Input | Outcome | Spearman | Interpretation |
|---|
Particle population
Precipitation evolution
Physics coupling
Run fragmentation or precipitation.
Population physics
PSN–pinning regime & kinetic milestones
Preserves the measured distribution, separates count-, volume- and pinning-weighted sizes, and reports when precipitation nucleates, saturates and begins to coarsen.
—Run particle analysis
Pinning contribution by particle class
Mechanism interpretation
Run fragmentation or precipitation.
Microstructure chain
Risk and property screen
Best evaluated process candidates
| Score | Cold reduction | Anneal time | Strength | Grain size | Earing risk | Surface risk |
|---|
Robust process design
Feasible operating window, Pareto set & perturbation robustness
Replaces a single weighted winner with explicit constraints, nondominated candidates and stability against small process variation.
—Build process window
Feasible variable windows
| Variable | Min | P10 | Median | P90 | Max |
|---|
Robust recommended route
Dominant route drivers
| Input | Outcome | Spearman |
|---|
Event score across acquisition
Detected events
| Peak time | Score | Candidate mechanism | Layer | X | Y |
|---|
Evidence-aware event review
Feature attribution, threshold stability & labelled validation
Each event receives duration, dominant evidence, confidence and stability across nearby thresholds. A label column enables precision, recall and F1 reporting.
—Run event detection
Event evidence
| Time | Candidate | Duration | Dominant evidence | Confidence | Threshold stability |
|---|
Validation and next action
Run event detection.
Measured versus fitted
Residuals
Parameters, intervals and sensitivity
| Parameter | Estimate | 95% low | 95% high | Sensitivity |
|---|
Transferability check
Cross-validation, parameter stability & residual structure
Calibration quality is judged on held-out predictions, fold-to-fold parameter variation, residual bias, heteroscedasticity and autocorrelation—not calibration R² alone.
—Run cross-validation
Held-out prediction diagnostics
Parameter stability
| Parameter | Fold mean | SD | CV | Min | Max |
|---|
Validation outcome
Learning curve
Model competition
Selection uses cross-validation on the training partition only; final metrics remain held out.
| Model | CV score | Fold SD | Configuration | Status |
|---|
Permutation importance
| Feature | Importance | Repeat SD |
|---|
Data audit
Run an experiment.
Local explanation
Model-agnostic feature perturbation around one held-out observation.
Train a model first.
Microscopy machine learning
Interactive pixel/object learning
Train a lightweight random-forest segmentation model from SynchroMet labels or an image/label pair. Intensity, colour, local statistics, edges, texture proxies and spatial coordinates are used with held-out pixel validation and uncertainty maps.
Predicted classes
Model uncertainty
Per-class validation
| Class | Precision | Recall | F1 | Support |
|---|
Research dependency graph
Query results
| ID | Type | Label | Details |
|---|
Research authority
Provenance completeness & graph integrity
Checks orphan entities, broken relationships and missing core metadata before the graph is treated as a reproducible research record.
—Build graph
Build the graph first.
Shape Memory Alloys & Nitinol · Martensite Engineering Laboratory
Transformation crystallography, functional response and fatigue intelligence
One synchronized environment for B2→B19′ crystallography, PTMC/WLR, correspondence theory, cofactor and shear/shear compatibility, lattice metrology, variant selection, self-accommodation, transition-layer dislocations, cyclic constitutive response, elastocaloric screening and model handoff.
ReadyThe 50.00–51.00 at.% Ni reference database is loaded.
Shared material state
Changes here propagate through every analysis tab.
Custom B2/B19′ lattice
PTMC/WLR habit-plane solver
Solves correspondence variants, rank-one transformation twins and invariant-plane laminate solutions from the active lattice state.
Principal stretches and transformation
Habit-plane variants
| HPV | CV pair | Minor fraction | Twin shear | Habit normal B2 | IPS residual |
|---|
Correspondence theory, cofactor conditions and supercompatibility
Compares metric-correspondence compatibility with stretch/cofactor criteria and evaluates design targets.
CMC / SMC diagnostics
Supercompatible target designer
Measurement uncertainty propagation
B2→B19′ hard-sphere structure explorer
Evaluates the atom-contact construction and compares its lattice output with the active XRD-derived state.
Measured versus model lattice
Binary NiTi reference lattice at Ms
The embedded table preserves the reported B2/B19′ parameters and uncertainties. Interior CSS values are explicitly marked as endpoint-constrained interpolation, not measurements.
| Ni (at.%) | Ms (K) | aB2 (nm) | a / b / c B19′ (nm) | β (°) | CSS basis |
|---|
DSC cycle analyzer
CSV columns: time,temperature,heatFlow. Baseline, transformation intervals, hysteresis, enthalpy, entropy and dissipated-energy screens are calculated.
Refined XRD lattice-series analyzer
Import Pawley/Rietveld-refined lattice parameters with columns T,aB2,a,b,c,beta and optional u_* uncertainties.
Self-accommodation and HPVC intelligence
Combines calculated shape-strain cancellation with the measured 2/3/4/6-HPVC and critical-slip-stress trends.
Cluster compatibility
Interaction-work mechanism competition
Ranks reorientation and deformation-twinning trajectories across parent orientations under tension or compression.
Transition-layer incompatibility and TID slip selection
Scans B2 {110}<001> slip systems for the greatest reduction of a logarithmic-strain incompatibility indicator.
| Rank | Slip plane | Direction | Best shear | Incompatibility | LID-plane angle |
|---|
Microscopic habit-plane trace reconstruction
Triangulates a 3D interface normal from traces measured on two or more known section planes, then compares it with PTMC and invariant-line/eigenplane predictions.
One-dimensional thermomechanical cycle laboratory
Runs reproducible stress-, strain- and temperature-controlled paths with martensite-fraction tracking and explicit calibration warnings.
Functional-fatigue evolution
Runs a transparent, calibratable cycle-evolution law for transformation-temperature drift, residual strain, recoverable-strain loss and hysteresis change. This is separate from structural crack-life prediction.
Nitinol wire actuator
Elastocaloric material screen
SMA phase engineering, ML handoff and reproducibility
Ranks binary-NiTi candidates, packages crystallographic results and creates handoffs for independent PTMC, OpenSees and native constitutive implementations.
Candidate ranking
Export and solver handoff
For full 3D thermomechanical behavior, use the exported parameters with a validated SMA UMAT or custom MOOSE/OpenSees material and benchmark against cyclic experiments.
Native open-source solver
VPSC
LANL VPSC mean-field polycrystal plasticity and texture evolution.
Upstream: official repositoryLicence: BSD-3-ClauseMatAMet stages and supervises the installed executable; upstream code is not silently replaced.
Prepared job
Upload a complete native case. Preflight validates the structure without requiring an installed executable; Run then invokes the configured binary.
Required native inputs
Files are not silently converted between solver conventions.
- Complete VPSC working directory
- Uppercase VPSC8.IN control file
- All texture, single-crystal, loading and auxiliary files referenced by the deck
Engine status
Executable, version, resource ceiling and command configuration.
Checking availability…
Job monitor
Live log, cancellation, result summary and files.
No job submitted in this browser session.
Queued
Execution rule: preflight can validate a deck without the executable. A reported numerical result requires the named native engine, a successful process exit and retained output files.
Native open-source solver
DAMASK Grid
DAMASK spectral grid solver for voxel-resolved crystal plasticity.
Upstream: official repositoryLicence: AGPL-3.0-onlyMatAMet stages and supervises the installed executable; upstream code is not silently replaced.
Prepared job
Upload a complete native case. Preflight validates the structure without requiring an installed executable; Run then invokes the configured binary.
Required native inputs
Files are not silently converted between solver conventions.
- material.yaml with phase and material sections
- load.yaml with solver/loadstep definitions
- Voxel geometry in VTK ImageData (.vti) with cell data
Engine status
Executable, version, resource ceiling and command configuration.
Checking availability…
Job monitor
Live log, cancellation, result summary and files.
No job submitted in this browser session.
Queued
Execution rule: preflight can validate a deck without the executable. A reported numerical result requires the named native engine, a successful process exit and retained output files.
Native open-source solver
Fierro LS-EVPFFT
LANL large-strain EVPFFT for local fields, grain rotation and texture evolution.
Upstream: official repositoryLicence: BSD-3-ClauseMatAMet stages and supervises the installed executable; upstream code is not silently replaced.
Prepared job
Upload a complete native case. Preflight validates the structure without requiring an installed executable; Run then invokes the configured binary.
Required native inputs
Files are not silently converted between solver conventions.
- Version-matched LS-EVPFFT primary input
- Voxel microstructure and orientation/material files referenced by that input
- Fierro build matched to the host CPU/GPU and MPI stack
Engine status
Executable, version, resource ceiling and command configuration.
Checking availability…
Job monitor
Live log, cancellation, result summary and files.
No job submitted in this browser session.
Queued
Execution rule: preflight can validate a deck without the executable. A reported numerical result requires the named native engine, a successful process exit and retained output files.
Native open-source solver
CPFEM
Finite-element crystal plasticity through a configured MOOSE-compatible solver.
Upstream: official repositoryLicence: LGPL-2.1MatAMet stages and supervises the installed executable; upstream code is not silently replaced.
Prepared job
Upload a complete native case. Preflight validates the structure without requiring an installed executable; Run then invokes the configured binary.
Required native inputs
Files are not silently converted between solver conventions.
- MOOSE-compatible .i input deck
- Mesh, Euler-angle, slip-system and parameter files referenced by the deck
- A crystal-plasticity-enabled MOOSE application executable
Engine status
Executable, version, resource ceiling and command configuration.
Checking availability…
Job monitor
Live log, cancellation, result summary and files.
No job submitted in this browser session.
Queued
Execution rule: preflight can validate a deck without the executable. A reported numerical result requires the named native engine, a successful process exit and retained output files.