Skip to main navigation Skip to search Skip to main content

Testing the reliability of fast SEM techniques for heavy mineral identification

  • Emily Finzel
  • , He Feng
  • , Michael Sinnwell
  • University of Iowa

Research output: Contribution to conference › Abstract › peer-review

Abstract

Heavy minerals, defined by densities greater than 2.8–2.9 g/cm³, typically comprise about 1% of siliciclastic sediments. Despite their low abundance, they are critical for provenance analysis, offering insights into weathering, sediment dispersal, tectonics, paleogeography, and recycling. This study evaluates the accuracy of rapid heavy mineral identification using a field emission scanning electron microscope (FE-SEM) equipped with dual energy-dispersive X-ray spectrometers (EDS) and AztecFeature software (Oxford Instruments). Two sample sets were used: one with 20–30 known heavy mineral grain types, and another with known mixtures. Mineral identification begins with a backscatter electron (BSE) image to distinguish particles from the epoxy background. Phase maps are then collected, and multi-phase particles are automatically split into distinct features. Only phase domains larger than 30 µm (coarse silt) are analyzed. EDS mapping is performed in two passes: a fast scan (1–3 µs dwell) to locate particles, followed by a slower scan (25–30 µs dwell) focused only on detected particles, significantly reducing acquisition time in samples with a high background-to-particle ratio. Simultaneously, morphological analysis captures key metrics such as area, aspect ratio, breadth, equivalent circular diameter, length, perimeter, and particle shape. Particles are initially classified using the default AztecGeo mineral database based on their chemical composition. For particles that remain unclassified due to compositional deviations from the database, post-processing is performed using GrainAlyzer software. A single mount containing hundreds to thousands of grains can be analyzed in 60–90 minutes. The workflow used in this study demonstrates that automated heavy mineral analysis via rapid elemental X-ray mapping combined with BSE imaging is both efficient and effective. It allows for direct mineral phase identification and supports the measurement of key physical attributes—such as size and shape—that are essential for thorough provenance analysis.
Original languageUndefined/Unknown
DOIs
StatePublished - 2025
Externally publishedYes

Cite this