Key Takeaways

  • TIMA, MLA, and QEMSCAN are all SEM-based automated mineralogy platforms — they scan a polished sample, fire an electron beam, and classify every mineral grain from its X-ray spectrum, at rates up to ~100,000 grains/hour.
  • All three now lean on machine-learning classifiers under the hood for particle/phase identification, not just fixed spectral lookup tables.
  • MLA is the original platform (Thermo Fisher/FEI); QEMSCAN and Zeiss Mineralogic are close alternatives, with Mineralogic explicitly marketing AI-based deep-learning classification.
  • TESCAN TIMA is the newer entrant and increasingly the default choice for high-throughput exploration and geometallurgy programs.
  • Output feeds directly into liberation analysis, grain-size distribution, and the clay/hardness/recovery ML models used elsewhere in your geometallurgy workflow.

TL;DR

Send polished sample mounts through an automated SEM mineralogy platform (TIMA, MLA, Mineralogic, or QEMSCAN) rather than manual point-counting — the instrument’s built-in ML classifier turns raw X-ray spectra into quantitative mineral, liberation, and grain-size data in hours instead of weeks.

How Do I Automate Mineralogy With AI?

Automated mineralogy replaces a human at a petrographic microscope doing manual point counts with a scanning electron microscope that rasters across a polished thin section or grain mount, collects an energy-dispersive X-ray (EDS) spectrum at each point, and classifies the mineral phase automatically. The “AI” part isn’t a bolt-on — it’s the core of how modern instruments work: TESCAN TIMA and Zeiss Mineralogic both use trained classification models (Mineralogic explicitly markets “AI-based deep learning algorithms”) to resolve ambiguous or mixed-pixel spectra that older lookup-table approaches would misclassify.

In practice, you’re choosing between three closely related instrument families rather than picking “the AI tool” versus “the manual tool” — Thermo Fisher’s MLA is the original mineral liberation analyzer and still widely deployed; QEMSCAN (originally CSIRO, now under Zeiss) and Mineralogic are the current Zeiss-branded successors; TIMA is TESCAN’s competing platform and has become a common default for new exploration and geometallurgy labs because of its multi-EDS-detector throughput. Which one you use is mostly a function of what your lab or contract lab already has installed — the workflow and outputs (modal mineralogy, liberation curves, grain-size distributions, elemental deportment) are functionally equivalent across all three.

Once you have quantitative mineralogy for a set of samples, that data becomes the feature set for the downstream ML models covered elsewhere in this series — clay-type classification, comminution hardness (BWI/DWI) proxies, and recovery prediction all consume automated-mineralogy output as their primary input. If you don’t already have in-house SEM capacity, commercial labs (ALS, SGS, Bureau Veritas) run TIMA/MLA/QEMSCAN as a standard service line, so this is accessible without a capital instrument purchase.

Try It With Geocluster

Automated mineralogy output is dense, multi-variable, and only useful once it’s joined against your geological, geotechnical, and metallurgical databases — which is exactly the kind of research and integration work Geocluster is designed to support. If you’re building geometallurgy models on top of TIMA/MLA/QEMSCAN data, Geocluster helps you connect that mineralogy to everything else in your dataset.