InterPore2026

Advanced Micro-CT Techniques for Visualizing Pore-Scale Microplastic retention patterns in Soils

21 May 2026, 15:35

1h 30m

Poster Presentation (MS10) Advances in imaging porous media: techniques, software and case studies Poster

Speaker

Marjan Ashrafizadeh (Institute for Geosciences, Applied Geology, Friedrich-Schiller-University Jena, 07749 Jena, Germany)

Description

The continuous release of microplastics (MPs <5 mm) has made them a global environmental concern. While early research focused on marine systems [1, 2], growing evidence shows that soils are a far larger and more persistent sink for plastic pollution [1, 3, 4]. MPs can also act as carriers for antibiotics, heavy metals, and organic pollutants, increasing risks to soil health, food safety, and groundwater quality [3, 5, 6]. Advances in X-ray computed tomography (CT) have enabled non-destructive visualization of large MP fragments in water, soils, and sediments. Previous studies have used CT to detect manually mixed MPs, image millimetre-scale plastic particles in organic-rich soils or assess how they alter soil pore structure and water-holding capacity [3, 6-9]. However, these studies rely on static systems where relatively large particles (>150 µm) were introduced manually, rather than capturing dynamic injection, flow, and transport processes.
In this study, we use high-resolution micro-CT (µCT) to visualize the transport and retention of 2-µm polystyrene MPs in soil columns (3 mm dimeter, 10 mm length) at low concentration (0.05 wt. %). After MP injection, soil columns were scanned at both saturated and dried conditions with a ZEISS Xradia 620 Versa system. A series of test scans with varying resolutions and acquisition parameters (e.g., fast versus high quality scans) were performed to optimize MP contrast in soil and minimize imaging artifacts. A multi-resolution scanning approach with voxel sizes ranged from 0.7 µm (the practical resolution limit of the scanner) to 8 µm was used. In addition to conventional absorption-contrast imaging, propagation-based phase-contrast imaging was employed to enhance MP visibility. Furthermore, multi-position vertical scan stitching was also used to capture the full column length. Reconstruction was performed using filtered back-projection followed by a non-local means filtration. A multiphase Random Forest-based segmentation algorithm was implemented to segment soil and pore spaces and to detect microplastics (Figure 1). All imaging configurations, including different voxel sizes, contrast modes, and scanning geometries, were systematically compared to identify the most effective micro-CT strategy for revealing the distribution of MPs in soil columns.
Although the smallest voxel size (0.7 µm) produced the clearest images, each scan required long scanning hours (>80 hours) and covered only a small field of view of the column, making it unsuitable for full-column studies. Phase-contrast imaging at this resolution was also highly time-intensive (approximately 34 hours per scan), but it produced the most noise-free images and the strongest MP contrast. After evaluating all options, we selected a voxel size of 2 µm as the best compromise between resolution, contrast, scan time, and field of view for MPs used in our study. This resolution allowed us to detect individual MPs and small clusters while imaging the full soil column length. To achieve this, the source-to-sample distance was minimized, and detector settings were optimized for each scan. Four vertically stacked scans with ~14% overlap were acquired to image the entire column, and the reconstructed volumes were stitched into a final dataset of ~1,000 × 1,000 × 3,500 voxels.