Can CT Imaging be Used to Investigate Microplastics in Sediments?

Abstract

https://doi.org/10.1016/j.marpolbul.2023.115350 Introduction: X-ray Computed Tomography (CT) is a three-dimensional non-destructive imaging technique, which could in theory allow for the non-invasive identification of microplastics in-situ within sediment cores. In this study, the utility of CT was explored as a novel technique for the non-destructive analysis of microplastics in river sediments, forming part of a wider study investigating sources and sinks of microplastics in the river Thames estuary, UK. Methods: The technique was tested in three stages: (1) layered cores (for assessing detectability), (2) randomly spiked cores (simulating environmental samples), and (3) environmental samples (real-world examples). Layered and spiked cores were created using a stock of homogenised intertidal foreshore sediment collected from Purfleet on the Thames estuary. Each of the four cores for both types were spiked with a different plastic polymer to test the influence of plastic type on recoverability. They were also categorised as ‘large’ (~4 mm diameter) or ‘small’ microplastics (~125 μm diameter), to assess the impact of size on recoverability. For the environmental samples, two sediment gravity cores were used which had been extracted from a transect of Barking Creek (a tributary of the Thames) as part of the wider study.  Results: When examined in layers within artificial cores, all microplastic types could be observed by CT imagery, with good contrast in X-ray attenuation against background sediments. The size and shape were reliably represented, allowing discrete particles to be classified using existing typology (Fig. 1). Differences in attenuation between polymer types suggested that it may be possible to uniquely identify different plastic polymers. Large microplastics were also detectable within spiked cores, having sufficient difference in attenuation to isolate individual microplastics (Fig. 2). Due to limitations on scan resolution, smaller microplastics could not be detected in spiked cores. Imaging of environmental samples revealed two distinct cores in terms of sediment structure, and a range of particle types were visible in the imagery of both (Fig. 3). It was not yet possible to confidently differentiate microplastic particles from others within these cores. Figure 1: Grayscale CT image slices of layered cores, annotated with image average gray level values for different particle types. Figure 2: False colour 3D reconstruction of polypropylene (left) and PET (right) spiked core scans, when suspected microplastics were isolated by a specified grayscale colour range and digitally coloured according to particle volume.  Figure 3: CT image slices of environmental cores UT3C (top) and UT3F (bottom). Core UT3C was extracted from Barking Creek (London, UK) and spanned depths of 0-30 cm, while UT3F was from the adjacent saltmarsh, spanning depths of 30-60 cm. Conclusions: This study shows that microplastics can be detected in sediments using CT. As a highly complementary approach, it has potential to be used alongside established techniques to reveal additional information about the sediment structure. While further work is needed to make it a robust, reproducible screening technique for the direct investigation of microplastics in environmental samples, this is an important early step in the use of CT within the field.