Execute predefined workflows
- Automatic sample loading, changing and measurement
- Acquire data using preset methods
- Save and transfer results automatically
AI-powered autonomous laboratories
Move XRD beyond automatic measurement
toward autonomous experiments.
Who analyzes the measurements? Are the results valid? What should be measured next?
iPowder provides intelligent XRD analysis and evaluation to bring structural information into experimental decisions.
Implemented · XRD instrument connection Project-specific · Materials rules and system integration
Connection and control of GKINST XRD equipment are implemented. Combine this link with robots, scheduling and data systems to build autonomous experimental loops.
01 / Autonomous laboratories
After automated measurement,
analysis and evaluation are still needed.
Automation executes predefined operations. Autonomous experiments must also interpret results and choose subsequent actions.
Turn results into experimental evidence
- Check data validity and signal-to-noise ratio
- Identify phase changes, unknown peaks and abnormal parameters
- Decide whether to remeasure, adjust conditions or screen samples
02 / Autonomous laboratories
iPowder’s role:
intelligent XRD analysis and decisions.
Experimental systems organize preparation and measurement; iPowder converts patterns into phase, cell, refinement, stress and structural information.
System overview; interfaces and scheduling depend on the actual laboratory configuration.
- Experiment design AI Scientist / Experimental goals
- Sample preparation Robot / Synthesis system
- Automatic sample loading Associate samples and tasks
- XRD instrument Measurement & Data acquisition
iPowder AI
Scientific analysis · Result evaluation
Select workflows, call algorithms and check intermediate results automatically.
Phases / Phase fractions Unit cells / Refinement Stress / Structural information
Evaluate Data quality, target status and anomalies
→Decide Remeasure, adjust conditions or continue screening
Feedback to the execution system
Instruments, robots and schedulers execute the next experiment according to configured strategies.
The instrument measures; iPowder interprets what was measured.
03 / Autonomous laboratories
Connect automatic analysis
with the next experiment.
iPowder receives data, selects methods and evaluates results. Complete experimental feedback requires scheduling and device-control integration.
Workflow illustration; evaluation rules and strategies must be configured for the materials system.
Receive experimental data automatically
- Monitor data and identify tasks
- Link sample information and check completeness
- Trigger analysis after measurement
Select methods for the task
- Phases: check → find peaks → search → evaluate
- Mixtures: search → structures → refine → quantify
- Batches: process → extract → trends → anomalies
Check results and recommend actions
- Data validity and signal-to-noise ratio
- Unknown peaks, missing phases and convergence
- Abnormal parameters and the need for further experiments
- First measurement
Target phase 40%
Analysis reveals residual precursors and informs changes to experimental conditions.
- Evaluate & Decide
Adjust reaction conditions
Example strategy: raise the reaction temperature, then let the experimental system repeat synthesis and measurement.
- Second measurement
Target phase 75%
Reanalyze composition and residuals to decide whether further optimization is required.
↻ Results feed back into strategy for continued iteration.
- SynthesizeSynthesize
- MeasureMeasure
- AnalyzeAnalyze
- EvaluateEvaluate
- DecideDecide
- OptimizeOptimize
- RepeatExperiment again
04 / Autonomous laboratories
Coordinate layers
for strategy, computation and control.
AI Agents plan analysis and experiments, scientific modules compute results, and scheduling and device systems execute tasks.
Architecture illustration; iPowder provides scientific analysis and collaborates with instruments and intelligence layers.
Top layer · AI Agent
Planning & Decisions Understand goals and plan tasks
Interpret experimental goals, call scientific modules, combine results and propose the next experimental strategy.
Tasks and analysis goals ↓ ↑ Results and diagnostics
Middle layer · iPowder scientific analysis
Scientific computation Process data and calculate structural information
Phases, Rietveld refinement, quantification, stress, structural parameters and data-quality assessment.
Scheduling and control requests ↓ ↑ Measurement data and device status
Bottom layer · Instruments & Control systems
Device execution Execute commands and safety interlocks
Device controllers operate motors, sources, detectors and sample changers, enforcing equipment constraints and safety interlocks.
Device controls enforce constraints, scientific algorithms compute evidence, and AI Agents plan subsequent actions.
05 / Autonomous laboratories
Implemented:
the link between XRD instruments and analysis.
iPowder connects seamlessly with GKINST XRD equipment, linking instrument measurement with data analysis and control.
▶ Watch software demonstration An existing iPowder recording shows analysis operations; it is not a field recording of instrument integration.
An implemented connection
From standalone analysis to integrated software, instruments and AI.
- XRD instrument interface Connection & Control
- Automatic data transfer Feed measured data into analysis
- Task triggers Initiate analysis workflows
- iPowder automatic analysis Scientific algorithms & Agents
- Analysis feedback Return to task and experiment systems
Instrument-to-analysis chain
Connection, data transfer, analysis triggering and feedback for the described XRD equipment.
Materials & Experimental systems
Robots, sample changers, LIMS, MES, scheduling, materials rules and experimental feedback logic.
Autonomous experimental loop
Use supporting systems and adapted rules for evaluation, optimization and execution over multiple rounds.
06 / Autonomous laboratories
Configure for materials and tasks.
Support real experimental applications.
Different materials need different structural information and feedback. Configure analysis rules, databases, thresholds, refinement strategies and feedback logic.
Phase transitions & Time evolution
Phase transitions, cell changes, cycling and in situ time-series analysis.
Structural changes during reactions
Phase transformations, intermediates and temperature dependence.
Relate phase composition to engineering properties
Phase fractions, lattice parameters, residual stress and texture.
Project configuration Analysis rules Databases Decision thresholds Refinement strategies Experimental feedback logic
Autonomous laboratories require configurable integration and project-specific adaptation for their materials, equipment and goals.
Find candidate materials among many samples
Following preparation and measurement, extract phases, fractions and cell parameters, then screen against configured objectives.
Batch measurement → Automatic analysis → Parameter comparison → Screening
Track continuously changing experiments
Monitor new phases, transformations, unit cells and reaction progress under changing temperature, electrochemical or atmospheric conditions.
Continuous patterns → Time-series analysis → Change evaluation → Adjust conditions
Compare conditions against target outcomes
For example, compare structural results at 800, 850, 900 and 950 °C to assess which conditions are closest to the target.
Change conditions → Compare structures → Evaluate targets → Recommend optimization
Connect abnormal results with remeasurement and alerts
Apply product rules and thresholds, then route anomalies to remeasurement or alerts through the experimental system.
Measure → Analyze → Check acceptance → Remeasure / Alert
07 / Autonomous laboratories
Work with experimental systems
to advance autonomous materials laboratories.
iPowder interprets X-ray characterization data. A complete loop also needs instruments, sample handling, scheduling and data systems.
Interface illustration; data and task flows do not imply that all interfaces are enabled by default.
Instrument layer XRD / Instrument control system Other characterization systems such as SAXS and XAS are directions for extension and adaptation.
↓ Data & Status Scientific analysis layer iPowderProcessing, structural information, quality assessment and results
↕ Tasks & Results Intelligence layer AI Agent / AI ScientistGoals, planning, evaluation and optimization strategies
↕ Scheduling & Feedback Automation & Management layer Robots / Schedulers / LIMS / MES / Databases Execution, sample management, process records and archives
Experimental system inputs
Tasks, sample information, measurement conditions, raw data and device status.
iPowder outputs
Phases, structural parameters, quality metrics, anomalies and recommendations to remeasure or adjust conditions.
Scientific algorithms Turn data into structural information Automation Execute analysis workflows continuously AI AgentOrganize analysis and evaluation strategies Open interfaces Connect equipment and experimental systems
Complete workflow
Design → Synthesize → Characterize → Analyze → Decide → Optimize
Connect automatic measurement, analysis and AI-assisted decisions into autonomous experiments.iPowder · Autonomous Laboratory
Make XRD an intelligent characterization node
in autonomous experimental loops.
Build on implemented instrument links and configure complete chains for materials, devices and tasks.