While SEM-EDS and automated mineralogy offer quick surface-based insights, they face several limitations:
Grain size, sphericity, and internal distribution cannot be captured from a single plane.
Minerals may go unnoticed if they’re not intersected by the prepared section.
- Sampling Bias from Density Settling
In resin mounts, heavier minerals like galena or gold may sink during curing, distorting abundance data.
- Poor Phase Contrast in Conventional CT
Minerals with similar attenuation, like pyrite and chalcopyrite, are indistinguishable, limiting analysis.
These issues compromise both the accuracy and representativeness of mineralogical data used for resource evaluation.
The Benefits of Using Spectral Micro-CT
Resolve Overlapping Attenuation Profiles
Spectral CT differentiates minerals like pyrite and chalcopyrite using complete X-ray attenuation curves instead of averaged grey levels.
Quantify Low-Concentration Minerals in 3D
Assess volume fractions of gold, tantalite, scheelite, or galena inside entire core samples or grain mounts without destructive prep.
Correct Sampling Bias from Resin Mounting
Visualize particle density gradients across the sample and retrieve accurate mineral distributions regardless of settling.
Enable Elemental Detection Using K-Edge Imaging
Identify heavy elements like Pb, W, Ta, or Au based on their absorption edges in the X-ray spectrum.