Visual Similarity Does Not Imply Equivalent Microplastic and Nanoplastic Burden: Optical Differentiation of Tap and Filtered Water

Microplastics and nanoplastics (MNPs) are increasingly recognized as environmental and potential human health concerns, with emerging evidence of widespread exposure through water sources. However, assessment of MNP burden remains challenging, particularly when samples appear visually indistinguishable.

In this report, tap water, filtered water, and double-filtered water were evaluated using top-down optical imaging at standardized timepoints. Across all conditions, samples appeared visually clear to the naked eye. Despite this, consistent differences in optical structure were observed, including variations in background haze, radial clearing, and spatial uniformity.

Notably, visually similar samples exhibited distinct optical signatures, supporting the principle that visual inspection alone is insufficient to assess MNP burden.

Figure 1. Grayscale optical comparison of tap, filtered, and double-filtered water at 30 minutes. Top-down images captured without a grid reference and converted to grayscale for visualization of spatial structure. While all samples appear visually similar to the naked eye, subtle but reproducible differences in background haze, radial clearing, and field uniformity are observed. These differences reflect underlying variation in microplastic/nanoplastic and dissolved-phase composition and are not apparent through visual inspection alone.

Computational analysis was performed using computer vision approaches that evaluate spatial structure and texture rather than detecting individual particles. This approach captures emergent optical patterns arising from particle interactions, enabling differentiation between samples without reliance on discrete particle identification.

Importantly, this framework is sensitive to nanoplastic contributions (<200 nanometer) that are difficult to resolve using conventional lab-based methods. In addition, consistent optical behavior was observed in salt water conditions, suggesting robustness of the interaction-driven detection approach in ionic environments that may deviate from classical colloidal assumptions. Consistent optical behavior observed across both freshwater and saline conditions further indicates robustness of the interaction-driven approach in varied environmental matrices.

These findings demonstrate that optical and computational methods can reveal meaningful differences in microplastic and nanoplastic burden even when samples appear visually indistinguishable, supporting the development of field-deployable tools that go beyond conventional visual or turbidity-based assessment. This report is also available at:

https://doi.org/10.5281/zenodo.19329352